|
| HOME | SUBSCRIPTIONS | CURRENT ISSUE | PAST ISSUES | CARDIOSOURCE | SEARCH | HELP | FEEDBACK |
|
|
|
|||||||||
|
|||||||||||
|
J Am Coll Cardiol, 2007; 49:951-959, doi:10.1016/j.jacc.2006.10.065
(Published online 16 February 2007). © 2007 by the American College of Cardiology Foundation |
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
Department of Cardiology, Toyohashi Heart Center, Toyohashi, Japan.
Manuscript received June 26, 2006; revised manuscript received October 23, 2006, accepted October 30, 2006.
* Reprint requests and correspondence: Dr. Mariko Ehara, Toyohashi Heart Center, Department of Cardiology, 21-1, Gobutori Oyamacho, Toyohashi-City, Aichi 441-8530, Japan. (Email: momomar{at}muc.biglobe.ne.jp).
 |
Abstract |
|---|
 Top Abstract MethodsResultsDiscussionSummary and ConclusionsReferences |
|---|
Background: A noninvasive detection of ISR would result in an easier and safer way to conduct patient follow-up.
Methods: We performed CTCA in 81 patients after stent implantation, and 125 stented lesions were scanned. Two sets of images were reconstructed with different types of convolution kernels. On CTCA, neointimal proliferation was visually evaluated according to luminal contrast attenuation inside the stent. Lesions were graded as follows: grade 1, none or slight neointimal proliferation; grade 2, neointimal proliferation with no significant stenosis (<50%); grade 3, neointimal proliferation with moderate stenosis (
50%); and grade 4, neointimal proliferation with severe stenosis (75%). Grades 3 and 4 were considered binary ISR. The diagnostic accuracy of CTCA compared with ICA was evaluated.
Results: By ICA, 24 ISRs were diagnosed. Sensitivity, specificity, positive predictive value, and negative predictive value were 92%, 81%, 54%, and 98% for the overall population, whereas values were 91%, 93%, 77%, and 98% when excluding unassessable segments (15 segments, 12%). For assessable segments, CTCA correctly diagnosed 20 of the 22 ISRs detected by ICA. Six lesions without ISR were overestimated as ISR by CTCA. As the grade of neointimal proliferation by CTCA increases, the median value of percent diameter stenosis increased linearly.
Conclusions: Binary ISR can be excluded with high probability by CTCA, with a moderate rate of false-positive results.
| |||||||||||
Recently 64-slice MSCT, the latest generation in current usage, has offered an improved spatial and temporal resolution. The accuracy of 64-slice MSCT to diagnose de novo coronary artery disease is also promising; however, segments after stenting were again excluded in most of the populations of previously published data.
The goal of the present study is to evaluate the diagnostic accuracy of coronary binary ISR with 64-slice MSCT compared with ICA. We also investigated the correlation of the visual estimation of ISR by CTCA with quantitative coronary angiography by ICA.
|
Methods |
|---|
|
TopAbstractMethodsResultsDiscussionSummary and ConclusionsReferences |
|---|
Scan protocol of CTCA. All patients were scanned on a 64-slice scanner (SOMATOM Sensation 64 Cardiac, Siemens Medical Solutions, Forchheim, Germany). When necessary, beta-blocker (metoprolol 20 to 60 mg) was administered for heart rate control. Use of nitroglycerin before scanning was left to the physicians’ discretion. A bolus of contrast media (iopamidol, 300 mg iodine/ml or 370 mg iodine/ml, Schering AG, Berlin, Germany; iopromide, 300 mg iodine/ml, Shering AG; Omnipaque, 300 mg iodine/ml or 350 mg iodine/ml, Daiichi Pharmaceutical Co., Ltd., Tokyo, Japan) was injected into an antecubital vein, followed by flushing with 20 to 50 ml saline. To reduce the incidence of adverse reaction, the sort of contrast media was selected for each patient considering previous usage. The proper amount of the contrast media and injection speed were determined according to the patient’s body weight, scan time, and heart rate. The start delay was automatically defined using bolus tracking software equipped in the scanner. The region of interest was placed within the ascending aorta, and the scan was started when the CT density reached 120 HU higher than the baseline CT density. The scan was performed between the tracheal bifurcation and diaphragm with the following parameters: collimation width 64 x 0.6 mm, rotation time 330 ms, tube voltage 120 kV, effective tube current 800 mA, table feed 11.5 mm/rotation, and pitch 0.2.
Data acquisition and image reconstruction of CTCA.
Image reconstruction was retrospectively gated to an electrocardiogram, and the optimal cardiac phase showing the minimum motion artifact was individually determined. Depending on the heart rate during the examination, axial slices were reconstructed synchronized to the electrocardiogram by a monophase (heart rate <70 beats/min) or biphase (heart rate 70 beats/min) reconstruction algorithm using data from 1 or 2 consecutive heart beats. When necessary, R-wave indicators were manually repositioned to improve the quality of synchronization. To optimize in some cases the image quality, different cardiac cycle phases were selected for different coronary segments.
As previously described (16), 2 sets of images were reconstructed with different types of convolution kernels: the first set was reconstructed with a smooth or medium-smooth kernel (B20f, B25f, or B30f), with a slice thickness of 0.75 mm (increment 0.4 mm), and the second set was reconstructed with a sharp (Heartview, Siemens Medical Solutions) kernel (B46f), with a slice thickness of 0.6 mm (increment 0.3 mm). Spatial resolution is 0.33 mm. For delineating low-contrast objects such as coronary lumen or vessel wall, we used images generated with the smooth or medium-smooth kernel, whereas to observe the stented segment, we used both of the images that were generated with smooth and sharp kernels. Figure 1 shows the difference between 2 types of convolution kernels.
|
ICA procedure and analysis.
The ICA was performed with standard techniques, and at least 2 different views were obtained for each main vessel. All stented segments (including the 5 mm proximal and distal to stent edges) were evaluated by a skilled observer who was blinded to the results of CTCA. Quantitative coronary angiography was performed, and percent diameter stenosis (%DS) was calculated using CMS (MEDIS, Leiden, the Netherlands), which was defined as reduction of minimal lumen diameter compared with the vessel diameter estimated with proximal and distal reference. Segments were classified into 4 groups according to %DS: Slight included segments with <25% in %DS, mild with 25 and <50%, moderate with 50 and <75%, and severe with 75%. Binary ISR was defined as %DS 50% (moderate or severe) in stented segment on ICA.
Validation, semiquantitative analysis, and morphological ISR classification on CTCA compared with ICA.
Referring to a report by Gasper et al. (14), contrast attenuation inside and at both edges of stents compared with the vessel lumen was regarded as neointimal proliferation. Adequately reconstructed CTCA images of stented segments were visually classified into 4 grades using the following criteria (Fig. 2): grade 1, none or slight neointimal proliferation; grade 2, mild neointimal proliferation but no significant restenosis (<50% narrowing); grade 3, moderate neointimal proliferation with significant restenosis (50% narrowing); grade 4, neointimal proliferation with severe stenosis or total occlusion (75% narrowing or occlusion). Two experienced observers who were blinded to the patients’ clinical information as well as to the ICA findings evaluated the grade of CTCA independently, and when the readings of the observers differed, a consensus was reached and used in the final analysis. All segments were blindly analyzed twice by each observer to evaluate the intraobserver reproducibility.
|
Statistical analysis. Quantitative variables are described as mean ± SD. Categorical variables are presented as numbers and percentages. Comparison of quantitative variables were performed by one-way analysis of variance for normally distributed variables. The chi-square test was used for comparing frequency of occurrence. Statview version 5.0 (Abacus Concepts Inc., Berkeley, California) was used for data analysis. A probability value of <0.05 was considered to indicate statistical significance. Interobserver and intraobserver agreements were expressed as Cohen kappa statistics. Sensitivity, specificity, positive predictive value (PPV), and negative predictive value (NPV) for CTA to detect significant stenosis were calculated from chi-square test of contingency, and 95% confidence intervals (CIs) were calculated from binomial expression.
|
Results |
|---|
|
TopAbstractMethodsResultsDiscussionSummary and ConclusionsReferences |
|---|
|
3 mm. There were 38 lesions (30%) with 2 overlapping stents, but no stents deployed inside other stents (Table 2).
|
0.005 inches) versus thin (<0.005 inches) was 86 (53%) versus 77 (47%) (Table 2). One hundred and fifty-eight (97%) stents were made of 316L stainless steel, 1 cobalt chrome, 2 nitinol, and 2 tantalum. The population had 33 (20%) drug-eluting stents (all Cypher stents, Cordis Corp., Miami, Florida) and 130 (80%) bare-metal stents. Binary ISR diagnosis: CTCA compared with ICA. Binary ISR was observed on ICA in 24 stented segments (19%) in 20 patients. The CTCA image quality was adequate in 110 stented segments (71 patients) of the 125 stented segments (81 patients). In these 110 assessable segments, 22 binary ISRs were diagnosed in 19 patients on ICA.
Of these 22 binary ISRs diagnosed on ICA, 20 (17 patients) were correctly detected by CTCA. Similarly, among 88 assessable lesions that had no ISR, 82 were correctly ruled out by CTCA (Table 3). To estimate the overall accuracy, including the 15 unassessable segments (2 with ISR and 13 without ISR on ICA), we assigned those segments as having binary ISR, because restenosis could not be excluded by CTCA. When including unassessable segments, sensitivity was 92% (22 of 24, 95% CI 81 to 100), specificity was 81% (82 of 101, 95% CI 74 to 89), PPV was 54% (22 of 41, 95% CI 39 to 69), and NPV was 98% (82 of 84, 95% CI 95 to 100). The predictive accuracy was 83%. When excluding unassessable segments, the sensitivity was 91% (20 of 22, 95% CI 79 to 100), specificity was 93% (82 of 88, 95% CI 88 to 98), PPV was 77% (20 of 26, 95% CI 61 to 93) and NPV was 98% (82 of 84, 95% CI 95 to 100). The predictive accuracy was 93%. The kappa-values of interobservers and intraobservers for the CTCA evaluation of stenosis were 0.85 and 0.92, respectively.
|
On the patient-based analysis, 10 patients had poor image quality in all of the stented segments. One of these patients had 2 binary ISRs, whereas the other 9 had no binary ISR. For the overall population, sensitivity was 90% (18 of 20), specificity 79% (48 of 61), PPV 58% (18 of 31), and NPV 96% (48 of 50). The predictive accuracy was 81%. When excluding unassessable patients, sensitivity and specificity of CTCA to detect binary ISR were 89% (17 of 19) and 92% (48 of 52), and PPV and NPV were 81% (17 of 21) and 96% (48 of 50). The predictive accuracy was 92% (Table 3).
Failure to detect binary ISR occurred in 4 lesions in 3 patients. Two ISRs in the same patient were missed because of unassessable CTCA images. A further 2 ISRs located in the left circumflex artery in 2 patients were missed by CTCA (false negative). In the first case, 2 overlapping stents (2 Cypher, 3.5 mm in diameter and 23 mm in length) were placed from the left main to the proximal left circumflex artery, in a severely calcified lesion. Binary ISR occurred in the bifurcation of the left main artery. In the second case, ISR in a single stent implanted in the proximal left circumflex artery (3 x 14 mm, DuraFlex, Avantec Vascular Corp., Sunnyvale, California) was not detected because of severe motion artifact. Concerning the 6 falsely detected ISRs, 2 were located in single stents of the proximal right coronary artery with motion artifact, 2 were in the proximal left ascending artery with moderate calcification (1 had overlapping site), 1 was in the mid-left ascending artery with no calcification but overlapping site, and 1 was in the mid-left circumflex artery with no calcification or overlapping but motion artifact as well.
Comparison between ICA and CTCA for the type of ISR based on the morphologic classification (Mehran classification) is shown in Table 4.
|
|
|
3.0- or 3.5-mm average diameter. Reduction of image quality was caused by extensive imaging artifacts generated by metallic struts themselves, motion artifact, calcification, or low opacification. In addition, in all unassessable stents, the CT density of stented lumens were <300 HU, which indicates lower opacification. Moderate to severe calcification was observed in 9 of those lesions. Stents implanted were 2 Cypher, 2 Bx Velocity (Cordis Corp.), 2 Express 2 (Boston Scientific, Maple Grove, Minnesota), 2 NIR (Boston Scientific, Galway, Ireland), 2 Wiktor (Medtronic Inc., Minneapolis, Minnesota), 1 DuraFlex, 1 MultiLink Penta (Abbott Vascular, Temecula, California), 1 S660, 1 Palmaz-Schuz Spiral (Johnson & Johnson Interventional System, New Brunswick, New Jersey), and 1 MultiLink Tristar (Abbott Vascular). |
Discussion |
|---|
|
TopAbstractMethodsResultsDiscussionSummary and ConclusionsReferences |
|---|
An ISR is the major long-term complication of percutaneous coronary treatment and can be divided into 2 main causes: neointimal proliferation and mechanical causes. Until now, only invasive methods were able to diagnose ISR (18,19). A noninvasive detection of ISR would be of clinical importance in the treatment and follow-up of coronary artery disease. The CTCA is less invasive and less expensive than ICA, which reduces physical, mental, and economical stress, as well as potential complications, especially for patients who require repeat coronary angiography. Because of a low ISR rate with drug-eluting stents, the indication of percutaneous coronary intervention widely extended to more challenging cases such as left main lesions, chronic total occlusion, bifurcations, long lesions, and small vessels. A routine follow-up in these at-risk populations with a reported ISR rate ranging from 8% to 25% may be preferred (2–4,20,21).
The most important features of a test for it to be useful in the clinical setting is its reliability of not missing the cases having the disease (sensitivity), in this case ISR, and achieving this with reasonably few falsely overdetected cases (specificity). Additionally, when the events searched for have a low occurrence rate, as is now the case for simple lesions treated with drug-eluting stents, a correct exclusion of restenosis (high NPV) allows focus on the small number of patients with a positive test result, in whom invasive angiography would be performed to confirm and treat the ISR, or definitively exclude it.
As for any imaging modalities, MSCT provides information about anatomy and lumen size, but does not provide information about whether a stenosis is flow limiting. However, nonsignificant in-stent neointimal proliferation is very unlikely to be flow limiting, allowing a noninvasive follow-up of these types of lesions. On the other hand, lesions classified as binary ISR will need further evaluation to clarify their pathophysiological significance, and this can be performed either noninvasively (i.e., myocardial perfusion scanning) or invasively (i.e., fractional flow reserve). Our results show that 64-MSCT correctly excludes binary ISR. However, the PPV for the overall population is low. As a result, nearly 1 of 2 patients sent for angiography after ISR detection by CTCA will have with a false-positive result.
It also should be emphasized that this study is a semiquantitative approach, and that a quantitative analysis probably would show a variation in the stenosis severity when expressed in absolute numbers. However, these variations in absolute stenosis severity between ICA and CTCA do not affect the binary decision for ISR, which is the most clinically relevant information. Additionally, the decision of how to treat ISR also depends on its pattern (focal, diffuse, proliferative, occlusive), which was adequately classified by CTCA in our study.
Based on this study, we think that 64-MSCT has a role to play in the detection of binary ISR. The best candidates for such an examination probably include patients with atypical symptoms. In these patients, a normal CTCA that excludes restenosis with a high probability would allow deferment of angiography. Asymptomatic patients should be followed up only clinically, and patients with typical symptoms should be referred directly for angiography.
The relatively high rate of ISR observed in our study is explained by the predominant use of bare-metal stents (82%), and also by a high proportion of treated complex lesions. Several studies reported the influences of stent morphology and stent diameter on the rate of ISR (22–24). Additionally, stent morphology influences its CT visualization as well as the underlying lumen, which may affect the ability of CTCA to detect ISR. In our study, no statistically significant differences according to stent morphology for the detection of ISR could be shown. This study was, however, not powered to answer such subanalysis, and differences according to stent morphology therefore cannot be excluded.
As mentioned earlier, causes of ISR can be divided mainly into 2 areas: neointimal proliferation and mechanical causes. The MSCT allows depicting finely the stent structure in a 2- and 3-dimensional way. An MSCT can detect stent fracture or deformity, as well as proper stent overlap and the effect of using the kissing balloon technique between the stent struts (25). These CT advantages became possible after improvement of the scanner’s spatial and temporal resolution (26,27).
Challenges in CTCA. Our analysis showed that CTCA had 6 false-positive segments. The main reasons for overestimation is calcification, overlapping of stents, and motion artifact. Similarly, we had 12% unassessable segments. Although in most part this problem is caused by inadequate resolution, we have some room for improvement.
Former MSCT generations were unable to depict fine objects with a high CT density, such as calcium or metallic structures, which generate artifacts. Therefore, many investigators had excluded candidates after stent implantation from their study population. Recently some investigators reported detection of ISR (13–16,28–32), and others presented in vitro stent depiction using MSCT (27,33). However, there are very few data on the visualization of in vivo stent structure. The improved resolution of 64-MSCT allows us to overcome most of those limitations.
To improve stent visibility and to decrease artifact, specific reconstructions were performed. In addition to reconstruction with ordinary field of view, phase of cardiac cycle, and convolution kernel (B20f or B30f, smooth or medium-smooth kernel), we made additional reconstructions stent by stent with another convolution kernel (B46f, HeartView kernel) in limited field of view (16). These reconstructions were repeated usually in a few different phases for each stent until a satisfactory image was obtained. A smooth kernel is suitable for delineating vessel lumen, vessel wall, and surrounding tissue, whereas a sharp kernel (HeartView kernel) is good at visualizing objects located next to fine and high-density obstacles such as calcium deposit or stent. The same reconstruction protocols were performed for each stented segment by both observers, with an excellent reproducibility, as shown by the low interobserver and intraobserver variability.
Moreover, we tried to get adequate heart rate control with beta-blocker administration if necessary, and with patient relaxation. Also, adequate examination explanation, rehearsal of breath-hold, and tools for body holding ensure better image quality, and consequently a better diagnostic accuracy. Examination preparation is particularly important in cases after stenting. The initial extra time spent for preparation or reconstruction allows saving a greater amount of time and energy for the analysis and interpretation of images with an adequate quality. One of the future challenges is to reduce the radiation doses, because it is currently higher than for ICA. On the other hand, the contrast media amount was similar to the amount used for ICA.
This study has some limitations. First, many different sort of stents were implanted, and the number of each stent was too small for proper comparison between stents. Second, we took approximately 8 months to collect the subjects, so that slight changes may have occurred in the opacification of contrast media, preparation of beta-blocker, or other procedure steps.
|
Summary and Conclusions |
|---|
|
TopAbstractMethodsResultsDiscussionSummary and ConclusionsReferences |
|---|
|
References |
|---|
|
TopAbstractMethodsResultsDiscussionSummary and ConclusionsReferences |
|---|
Related Article
This article has been cited by other articles:
|
|
 |
|
  D. A. Bluemke, S. Achenbach, M. Budoff, T. C. Gerber, B. Gersh, L. D. Hillis, W. G. Hundley, W. J. Manning, B. F. Printz, M. Stuber, et al. Noninvasive Coronary Artery Imaging: Magnetic Resonance Angiography and Multidetector Computed Tomography Angiography: A Scientific Statement From the American Heart Association Committee on Cardiovascular Imaging and Intervention of the Council on Cardiovascular Radiology and Intervention, and the Councils on Clinical Cardiology and Cardiovascular Disease in the Young Circulation, July 29, 2008; 118(5): 586 - 606. [Full Text] [PDF]
|
|
|
|
 |
|
 D. Oncel, G. Oncel, A. Tastan, and B. Tamci Evaluation of Coronary Stent Patency and In-Stent Restenosis with Dual-Source CT Coronary Angiography Without Heart Rate Control Am. J. Roentgenol., July 1, 2008; 191(1): 56 - 63. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 F Pugliese, A C Weustink, C Van Mieghem, F Alberghina, M Otsuka, W B Meijboom, N van Pelt, N R Mollet, F Cademartiri, G P Krestin, et al. Dual source coronary computed tomography angiography for detecting in-stent restenosis Heart, July 1, 2008; 94(7): 848 - 854. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
W T Roberts, J J Bax, and L C Davies Cardiac CT and CT coronary angiography: technology and application Heart, June 1, 2008; 94(6): 781 - 792. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 S. Schroeder, S. Achenbach, F. Bengel, C. Burgstahler, F. Cademartiri, P. de Feyter, R. George, P. Kaufmann, A. F. Kopp, J. Knuuti, et al. Cardiac computed tomography: indications, applications, limitations, and training requirements: Report of a Writing Group deployed by the Working Group Nuclear Cardiology and Cardiac CT of the European Society of Cardiology and the European Council of Nuclear Cardiology Eur. Heart J., February 2, 2008; 29(4): 531 - 556. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 A. N. DeMaria, J. J. Bax, O. Ben-Yehuda, P. Clopton, G. K. Feld, G. S. Ginsburg, B. H. Greenberg, J. D. Knoke, W. Y.W. Lew, J. A.C. Lima, et al. Highlights of the year in JACC 2007. J. Am. Coll. Cardiol., January 29, 2008; 51(4): 490 - 512. [Full Text] [PDF]
|
|
|
|
|
|
J. Abdulla, S. Z. Abildstrom, O. Gotzsche, E. Christensen, L. Kober, and C. Torp-Pedersen 64-multislice detector computed tomography coronary angiography as potential alternative to conventional coronary angiography: a systematic review and meta-analysis Eur. Heart J., December 2, 2007; 28(24): 3042 - 3050. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
B. S. Lewis and D. A. Halon Integrating Multidetector Computed Tomography Into Clinical Practice: Computed Tomography Scanning Shows its Metal J. Am. Coll. Cardiol., March 6, 2007; 49(9): 960 - 962. [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | SUBSCRIPTIONS | CURRENT ISSUE | PAST ISSUES | CARDIOSOURCE | SEARCH | HELP | FEEDBACK |
