| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
Original Research |
1 Department of Radiology, Sifa Hospital, Fevzipasa Blvd. 172/2, 35340 Basmane,
Izmir, Turkey.
2 Department of Cardiology, Sifa Hospital, Basmane, Izmir, Turkey.
Received December 18, 2007;
accepted after revision January 28, 2008.
Address correspondence to D. Oncel
(dilekoncel{at}hotmail.com).
Abstract
 Top Abstract IntroductionSubjects and MethodsResultsDiscussionReferences |
|---|
SUBJECTS AND METHODS. Thirty-five consecutively registered patients (10 women, 25 men; mean age, 65 years) with 48 stents were examined prospectively without heart rate controlling agents. Observers evaluating image quality and patency of the stents were blinded to the results of invasive coronary angiography. In-stent restenosis was defined as more than 50% narrowing of the lumen.
RESULTS. All stents were considered assessable for diagnosis with dual-source CT. In 85% (41/48) of the stents, image quality was good. Only two patent stents were misidentified as being stenotic. All other stents with stenosis and occlusion were correctly diagnosed. The sensitivity, specificity, positive and negative predictive values, and accuracy of dual-source CT in the detection of in-stent restenosis and occlusion were 100%, 94%, 89%, 100%, and 96%, respectively. The McNemar test result showed no statistically significant difference between the diagnostic performance of dual-source CT and that of invasive coronary angiography. The kappa indexes showed excellent intraobserver and interobserver agreement.
CONCLUSION. The high temporal resolution of dual-source CT is helpful for evaluation of coronary stents without heart rate control. Further confirmation of our preliminary results may broaden the clinical indications for CT angiography as a diagnostic test for the exclusion of in-stent restenosis.
Keywords: coronary CT angiography • coronary stents • dual-source CT
|
TopAbstractIntroductionSubjects and MethodsResultsDiscussionReferences |
|---|
|
TopAbstractIntroductionSubjects and MethodsResultsDiscussionReferences |
|---|
Patient Preparation
Irrespective of individual heart rate and heart rate variability, no
β-blockers were given before scanning. Sublingual nitrate (5 mg of
isosorbide dinitrate, Isordil, Fako) was given 5 minutes before image
acquisition to dilate the coronary arteries.
Scan Protocol
All CT examinations were performed on a dual-source CT scanner (Somatom
Definition, Siemens Medical Solutions). The coronary angiographic scan was
obtained with injection of 70 mL of nonionic contrast medium (400 mg I/mL
iomeprol, Iomeron, Bracco) at a flow rate of 6 mL/s followed by 50 mL of
saline solution (injection rate, 5 mL/s) to wash out the contrast material
from the right ventricle. Contrast administration was controlled with bolus
tracking. The scan parameters were detector collimation, 32 x 0.6 mm;
slice acquisition, 64 x 0.6 mm; gantry rotation time, 330 milliseconds
(temporal resolution, 83 milliseconds); pitch, 0.2–0.47 adapted to the
heart rate; tube current, 390 mAs per rotation; tube potential, 120 kV.
Scanning time was approximately 5.7–8.4 seconds, depending on the
cardiac dimensions and pitch, in a single breath-hold in the craniocaudal
direction. Pro spective ECG tube-current modulation (ECG pulsing) for
radiation dose reduction was used for all patients.
Image Reconstruction
Retrospective gating technique was used to synchronize data reconstruction
with the ECG signal. The reconstructions were made in all cardiac phases at
50-millisecond intervals at a slice thickness of 0.75 mm and a reconstruction
increment of 0.5 mm. The reconstruction interval with the fewest motion
artifacts was chosen and used for further analysis. To decrease stent-related
artifacts, edge-enhancing high-spatial-resolution kernels (B46f) were used for
reconstruction [16,
17].
To improve the delineation of the stents, the images were displayed in zoom mode at a window level of 200 H and window width ranging from 700 to 2,000 H. We found that the combination of a 200 H level with 1,500 H width was best for visualization of stents with respect to in-stent luminal dimension. The data obtained at this setting therefore were used for further analysis. The window settings also were optimized for soft-tissue visualization in the evaluation of intimal hyperplasia causing in-stent stenosis.
Noninvasive MDCT Angiographic Analysis
Analysis of scans was performed at a workstation (Wizard, Siemens Medical
Solutions) equipped with dedicated cardiac postprocessing software (Syngo
Circulation, Siemens Medical Solutions). The images were evaluated by two
independent radiologists with 5 years of cardiac CT experience. They were
unaware of the clinical data and were blinded to the results of invasive
coronary angiography. Disagreements were resolved by consensus, which was
achieved in all discordant cases.
The observers evaluated the data sets on both the original axial images and multiplanar reformatted reconstructions orthogonal and perpendicular to the vessel course. Curved multiplanar reformations were made both manually and with automated software. Contrast enhancement within the lumen of the stented segment was compared visually with enhancement in the unstented portion of the artery. Short-axis views were examined at various points along the stent, particularly where reduced luminal enhancement was identified.
The assessability of each stent was determined. A stent was considered assessable when the stent lumen was visible and contrast attenuation of the lumen could be evaluated qualitatively without the influence of partial volume effects, metal artifacts of stents, or cardiac motion artifacts. Each stent was assigned an image quality score of 1 (good image quality, no artifacts affecting evaluation of the stent), 2 (moderate image quality, mild to moderate artifacts, blurring but acceptable for clinical diagnosis), or 3 (poor image quality, uninterpretable with severe artifacts making stent evaluation impossible) according to the criteria used for assessability. The causes of reduced image quality were evaluated. Stent size and material also were considered.
Stents were visually evaluated and defined as 1, patent with no visible neointimal hyperplasia (absence of low-attenuation areas related to neointimal tissue); 2, patent with nonocclusive neointimal hyperplasia (longitudinal low-attenuation areas along the stent wall observed as a rim of hypoattenuation between the stent and the contrast-enhanced vessel lumen with residual lumen > 50%); 3, patent with in-stent restenosis (longitudinal and transverse low-attenuation areas along the stent wall with residual lumen > 50%); or 4, in-stent occlusion (complete loss of attenuation inside the stent lumen). The presence of persistent stenosis was defined as 50% or greater narrowing of the luminal diameter 5 mm proximal and distal to the stent. In-stent luminal diameter was measured manually with electronic calipers available at the workstation. Measurements were made on transverse images and on the curved multiplanar reformatted images, and the results were averaged. The location of the stents was specified as outlined in the American Heart Association 16-segment classification of coronary arteries [18]. Stent type and stent parameters were recorded.
The results of CT coronary angiography in the detection of in-stent
restenosis (lesions 
50%) and occlusion were compared with the results of
invasive coronary angiography according to a perstent analysis of each stent
individually and a perpatient analysis of the presence of a restenosed or
occluded stent in a given patient. The diagnostic performance of dual-source
CT coronary angiography in the evaluation of stents was assessed with respect
to image quality scores. Stent size and material were considered.
The percentage variability of the heart rate 
was calculated
according to the following equation:
 |
We evaluated the patient dose. The dose–length product was displayed by the dual-source CT system itself and converted into effective dose values with a conversion factor of 0.017 mSv/mGy·cm according to the Commission of the European Communities guidelines on quality criteria for CT [20].
Invasive Coronary Angiography: Reference Standard
Invasive coronary angiography was performed on all patients with standard
techniques 1 day after the CT examination. The angiograms were evaluated by
one experienced cardiologist (10 years of angiographic experience) blinded to
the results of dual-source CT angiography. The locations of the stents were
specified according to the guidelines of the American Heart Association
[18]. Restenosis was defined
as 50% or greater stenosis anywhere within the stent or within the 5-mm
segments proximal or distal to the stent margins. As in the CT evaluation,
stents were defined as patent with no intimal hyperplasia, patent with
neointimal hyperplasia (residual lumen > 50%), patent with in-stent
restenosis (residual lumen > 50%), or occluded. Patient-based analysis was
an evaluation of the presence of a restenosed or occluded stent in a given
patient. Patient dose was evaluated. The dose–area product displayed by
the system was converted into effective dose values with a conversion factor
of 0.183 mSv/mGy·cm2 according to the National Radiological
Protection Board guidelines
[21].
Statistical Analysis
Statistical analysis was performed with SPSS 12.0 (SPSS) for Microsoft
Windows. The diagnostic per formance of dual-source CT angiography in
evaluation of coronary stent restenosis and occlusion was determined with
respect to image quality scores and results of per-stent and per-patient
analyses. Sensitivity, specificity, positive and negative predictive values,
and accuracy were calculated. These diagnostic characteristics were expressed
with 95% CI. Invasive coronary angiography was the reference standard.
The McNemar test was used to determine whether there was a statistically significant difference between dual-source CT angiography and invasive coronary angiography in the evaluation of coronary stents. A value of p < 0.05 was considered statistically significant. To avoid bias caused by multiple stents per patient, p was adjusted with the correction factor C defined by Gonen et al. [22]. The results were checked with a generalized estimation equations methods program (SAS/STAT release 8.2, SAS Institute).
In the assessment of internal validity, all dual-source CT data sets were analyzed twice by two observers. The interval between the two analyses was 2 weeks with different orders of evaluation. Intraobserver and interobserver agreement on coronary stent evaluation was determined with kappa statistics. According to Landis and Koch [23], a kappa value of 0 indicates poor agreement; 0.01–0.20, slight agreement; 0.21–0.40, fair agreement; 0.41–0.60, moderate agreement; 0.61–0.80, good agreement; and 0.81–1.00, excellent agreement. Because of the small number of subjects, retrospective power analysis was performed.
|
TopAbstractIntroductionSubjects and MethodsResultsDiscussionReferences |
|---|
|
|
Dual-source CT and invasive coronary angiography were successfully performed on all patients without complications. The mean heart rate during CT angiographic scanning was 74.1 ± 12.1 beats/min (range, 56–98 beats/min). Twenty (57%) of the 35 patients had a heart rate less than 70 beats/min (mean, 64.7 ± 8.6 beats/min; range, 56–69 beats/min), and 15 (43%) had a heart rate greater than 70 beats/min (mean, 81.6 ± 4.3 beats/min; range, 73–98 beats/min). All patients were in sinus rhythm. The average percentage variability of heart rate was 6% ± 1.1% (range, 1.5–18.7%). The mean scanning time was 6.52 ± 0.71 seconds (range, 5.7–8.1 seconds). The mean scan pitch was 0.35 ± 0.08 (range, 0.24–0.47). The mean patient dose was 12.3 ± 1.52 mSv (range, 8.8–14.1 mSv).
|
|
|
|
|
|
At CT angiography, 28 of the 29 patent stents (Fig. 1A, 1B, 1C) with no intimal hyperplasia were correctly depicted. One patent stent was misidentified as stenotic. Eleven of 12 stents with in-stent restenosis (Fig. 2A, 2B, 2C) and one of two cases of neointimal hyperplasia (Fig. 3A, 3B, 3C) also were correctly diagnosed. One case of reste nosis was overestimated as occlusion and one case of neointimal hyperplasia as restenosis. Images of all of the misidentified stents were of moderate quality, and the stents were 3 mm in diameter or smaller. All five occluded stents were correctly identified (Fig. 4A, 4B, 4C, 4D). As at invasive angiography, no perisegment stenosis was detected.
|
|
|
|
|
|
|
According to these results, the sensitivity, specificity, positive and negative predictive values, and accuracy of dual-source CT in the detection of in-stent restenosis and occlusion were 100% (17/17), 94% (29/31; 95% CI, 85–100), 89% (17/19; 95% CI, 72–100), 100% (29/29), and 96% (46/48; 95% CI, 89–100) (Table 3). In patient-based analysis, two misdiagnoses of in-stent restenosis were made; in the other 33 patients, presence or absence of in-stent restenosis or occlusion was correctly diagnosed. The sensitivity, specificity, positive and negative predictive values, and accuracy of dual-source CT in the detection of in-stent restenosis and occlusion were 100% (16/16), 89% (17/19; 95% CI, 72–100%), 89% (16/18; 95% CI, 71–100%), 100% (17/17), and 94% (33/35; 95% CI, 84–100%), respectively (Table 3). On an intention to treat basis, 17 patients would have avoided invasive coronary angiography because in-stent restenosis and occlusion were excluded with CT coronary angiography. Two patients would have undergone invasive coronary angiography unnecessarily.
|
On good-quality images, all of the stents were correctly assessed. When
image quality was moderate, specificity, positive predictive value, and
accuracy decreased substantially (Table
3). The results of the McNemar test showed no statistically
significant difference between dual-source CT angiography and invasive
coronary angiography in the detection of in-stent restenosis and occlusion
(corrected p = 0.74). Results of analysis with generalized estimation
equations methods verified our results. Regarding image quality and diagnostic
performance, no significant difference was found between patients with low
(< 70 beats/min) and those with high ( 70 beats/min) heart rates
(p < 0.05).
Both interobserver and intraobserver agreement were excellent (
=
0.93 and = 0.96, respectively) for the detection of instent restenosis
and occlusion. Because of the small number of subjects, retrospective power
anal ysis was performed. The power of the study to compare dual-source CT
angiography and invasive coronary angiography in the detection of in-stent
restenosis and occlusion was calculated to be 0.77.
|
TopAbstractIntroductionSubjects and MethodsResultsDiscussionReferences |
|---|
The main improvement with dual-source CT is increased temporal resolution [14–16]. Early feasibility studies confirmed the technical capability of dual-source CT to produce diagnostic-quality images of the coronary arteries in patients with high heart rates [14–16, 29–31]. An in vitro study performed on vessel phantoms [30] showed that visualization of stenoses and stents was possible without motion artifacts at heart rates up to 120 beats/min. Image quality was similar for the static (0 beats/min) and dynamic (50–120 beats/min) scans. There are few clinical data, however, concerning the use of dualsource CT for the visualization of coronary stents. A study by Pugliese et al. [31] showed high diagnostic performance of dual-source CT in the detection of in-stent restenosis with no dependence on heart rate. In that study, only smaller stents (< 2.75 mm) caused diagnostic problems. Our results showed close correspondence with those findings.
A study by Scheffel et al. [29] showed high diagnostic image quality with dual-source CT without heart rate control in patients with extensive coronary calcifications. Because the spatial resolution of dual-source CT is the same as that of 64-MDCT, this difference in calcification dependency was explained by superimposition of the blooming artifact of severely calcified vessel walls with additional motion artifacts. This explanation leads to the conclusion that the high temporal resolution of dual-source CT may help to reduce artifacts caused by severe calcifications. We aimed to determine whether the improved temporal resolution may improve visualization of the coronary stents. Our results showed that dual-source CT performed well in determination of the presence of stent occlusion and in-stent restenosis. A negative correlation between heart rate and quality of coronary stent images that resulted in increased attenuation with increasing heart rate due to motion artifacts was previously reported [32].
In our study, all of the stents were assessable. Image quality was good in 85% (41/48) of the cases, and the assessments of those stents were correct. In all of the cases of moderate image quality, the stents were 3 mm in diameter or smaller. The reason for the decreased image quality was considered stent-related metallic artifacts, which were more disturbing owing to smaller sizes. Among those cases, one patent stent and one stent with nonocclusive intimal hyperplasia were misidentified as being stenotic, and one case of stenosis was misdiagnosed as occlusion. In one patient with two stents, wall calcifications contributed to degraded image quality.
Stainless steel and cobalt stents are better visualized [24, 29] than stents made with other materials. All of the stents in our study were stainless steel or cobalt, which produce fewer metal artifacts. In addition, most (67% 32/48) of the stents were larger than 3 mm in diameter. These factors most likely contributed to the high assessability rate and good image quality.
There was no correlation between poorer quality image or misdiagnosis and type of stent. There was no significant difference in the results for bare metal and drug-eluting stents. There also was no correlation with location of the stent in terms of right or left side of the circulation. Image quality and misdiagnoses, however, were strongly related to stent size. Images of all of the misidentified stents were of poor quality, and all of these stents were 3 mm or less in diameter and therefore located in small segments of the coronary vasculature.
No heart rate control was applied, and the heart rates ranged from 65 to 98 beats/min. Even in patients with high heart rates, diagnostic image quality was achieved. In the case of high heart rates, reconstruction intervals corresponding to the end-systolic phase yielded better image quality. In patients with heart rates less than 70 beats/min, reconstructions at the middiastolic phases were of better quality [19, 33].
While determining the presence of instent stenosis or occlusion, we used visual qualitative interpretation. In addition, the stent luminal diameters were measured manually with electronic calipers. The automated quantitative calculation function of the dedicated software was not used because the partial volume effects and blooming effect of the stent strut complicated the calculation.
We made no attempt to correlate the invasive coronary angiographic image projection with the evaluation CT projection to determine underestimation or overestimation of the stenosis in one plane on invasive coronary angiography. Invasive coronary angiographic examinations were performed with standard technique and standard projections by observers blinded to the results of CT coronary angiography. Similarly, observers evaluating CT coronary angiograms were blinded to the findings at invasive coronary angiographic examinations. In CT angiographic evaluations, stented segments were examined thoroughly in multiple orientations on curved multiplanar reconstruction images in both the long and short axes.
Prospective ECG-gated tube current modulation (ECG pulsing) limiting the
optimal radiation dose to a certain period of the cardiac cycle was used in
all patients to decrease the radiation dose
[34]. Because window width was
based on the mean heart rate during scanning, the dual-source CT system
allowed variable use of ECG pulsing at all heart rates. The mean radiation
dose in our study was 12.35 mSv with a range of 8.84–14.13 mSv. This
dose is comparable with the estimated dose (
13–18 mSv) for 64-MDCT
coronary angiography [35].
Because of heart rate adaptive table feed settings, the pitch also increased
at higher heart rates. Therefore, scanning time and patient dose decreased. In
contrast, at low heart rates, the adapted low pitch resulted in longer scan
times and a higher patient dose
[14–16,
29].
The main limitation of our study was that our patients were a high pretest probability population referred for invasive angiography with clinical suspicion of in-stent restenosis, which might have resulted in overestimation of the utility of dual-source CT for detecting and ruling out in-stent restenosis and occlusion. Wider applicability to a population with stents who have no symptoms may yield different results. Studies with larger patient populations are needed to confirm our initial experience.
Our results showed that the high temporal resolution of dual-source CT makes it helpful for visualization of the lumens of coronary stents without heart rate control. Confirmation of our preliminary results in larger patient populations may broaden the clinical indications for CT coronary angiography as a diagnostic test for the exclusion of in-stent restenosis, especially when the expected angiographic restenosis rate is low.
|
TopAbstractIntroductionSubjects and MethodsResultsDiscussionReferences |
|---|
CiteULike 
Complore 
Connotea 
Del.icio.us 
Digg 
Reddit 
Technorati What's this?
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
