HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS		QUICK SEARCH:   [advanced] Author: Keyword(s): Year:  Vol:  Page:

This Article Abstract Figures Only Full Text (PDF) All Versions of this Article: 2452061389v1 245/2/424    most recent Submit a response Alert me when this article is cited Alert me when eLetters are posted Alert me if a correction is posted Citation Map Services Email this article to a friend Similar articles in this journal Similar articles in PubMed Alert me to new issues of the journal Download to citation manager Citing Articles Citing Articles via HighWire Citing Articles via Google Scholar Google Scholar Articles by Das, K. M. Articles by Al Suwaidi, J. Search for Related Content PubMed PubMed Citation Articles by Das, K. M. Articles by Al Suwaidi, J.

Contrast-enhanced 64-Section Coronary Multidetector CT Angiography versus Conventional Coronary Angiography for Stent Assessment¹

¹ From the Departments of Radiology (K.M.D.), Cardiology (A.A.E., A.M.S., W.A.K.D., H.A.A., J.A.S.), and Medical Research (R.S.), Hamad Medical Corporation, Hamad Medical St, PO Box 3050, Doha, Qatar. Received August 10, 2006; revision requested October 13; revision received December 22; accepted January 15, 2007; final version accepted April 16. Address correspondence to K.M.D. (e-mail: daskmoy{at}gmail.com).

	ABSTRACT

TOP ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS DISCUSSION ADVANCES IN KNOWLEDGE IMPLICATIONS FOR PATIENT CARE... References

 
Purpose: To prospectively evaluate the accuracy of 64-section computed tomography (CT) for diagnosis of stent restenosis, by using conventional coronary angiography as the reference standard.

Materials and Methods: The ethics committee granted permission for the study; patients gave written consent. Contrast material–enhanced coronary CT angiography was performed in 53 patients (45 men, eight women; mean age, 54 years ± 9 [standard deviation]) suspected of having stent restenosis. Coronary CT angiographic findings were compared with conventional coronary angiographic findings. Two physicians analyzed coronary CT angiographic data sets with multiplanar reformatted images and three-dimensional reformations by using a volume-rendering technique and looked for stent detectability, low-attenuation in-stent filling defects, and grades of restenosis. Conventional coronary angiographic results were interpreted by one of several observers in consensus for stent restenosis; they were blinded to coronary CT angiographic data. Statistical software and general estimating equations were used for data analysis.

Results: One hundred ten stents were identified in 53 patients. Sensitivity, specificity, positive predictive value, negative predictive value, and accuracy of coronary CT angiography in detection of in-stent restenosis were 96.9%, 88.0%, 77.5%, 98.5%, and 91%, respectively. Coronary CT angiography depicted in-stent low-attenuation filling defects with an accuracy of 91% and negative predictive value of 98.5% (95% confidence interval: 90.9, 99.9). Coronary CT angiography depicted the status of 97 of 107 stents. There was no significant difference between in-stent lumen visibility and stent diameter (P = .104). Coronary CT angiography helped diagnose 15 of 18 stent restenoses with less than 50% narrowing, five of five stent restenoses with 50%–74% narrowing, and nine of nine (100%) stent restenoses with 75% or greater narrowing or total occlusion of the stent lumen.

Conclusion: Coronary CT angiography can depict in-stent low-attenuation filling defects, which appear to be a reliable sign of stent restenosis, and 64-section CT depicts such defects with a high degree of accuracy.

© RSNA, 2007

	INTRODUCTION

TOP ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS DISCUSSION ADVANCES IN KNOWLEDGE IMPLICATIONS FOR PATIENT CARE... References

 
Coronary stent restenosis has a reported incidence after 6 months that ranges from 11% to 46% in non–drug-eluting and of 0% in drug-eluting stents (1–3). Although 16-section multidetector computed tomography (CT) allowed improved spatial and temporal resolution in the elimination of stent-related high-attenuation artifact, the technique is still not adequate for clinical use (4–6). With 64-section CT, improved spatial and temporal resolution has been achieved, with better assessment of the coronary stent lumen (7). Thus, the purpose of our study was to prospectively evaluate the accuracy of 64-section CT for the diagnosis of stent restenosis, by using conventional coronary angiography as the reference standard.

	MATERIALS AND METHODS

TOP ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS DISCUSSION ADVANCES IN KNOWLEDGE IMPLICATIONS FOR PATIENT CARE... References

 
Patients
The study group consisted of 53 consecutive patients (45 men, eight women; mean age, 54 years ± 9 [standard deviation]) who underwent elective conventional coronary angiography for suspected stent restenosis from July 2005 to June 2006. Patients underwent conventional coronary angiography after contrast material–enhanced coronary CT angiography, and the average interval between these two examinations was 28 days ± 2. Coronary CT angiography was performed 25 months ± 14 after stent implantation. The ethics committee of the Hamad Medical Corporation granted permission for the study, and all patients gave written consent. Before signing the consent form, each patient was informed about the nature of the examination to be performed and associated hazards of radiation dose of each examination. The exclusion criteria included irregular heart rate, contraindication to iodinated contrast agents, or contraindication to ß-blocking drugs.

Multidetector CT Angiography
All patients were scanned with a 64-section CT scanner (Somatom Sensation 64; Siemens, Forchheim, Germany) with a 0.37-second rotation time. X-ray tube potential was 120 kV, with an effective tube current of 750–850 mA, 64 x 0.6 mm, table feed of 9.2 mm per rotation, and pitch of 0.24. Tube current modulation was not used to allow maximal flexibility in reconstruction intervals. Patients who were not taking ß-blocking drugs received 100 mg of atenolol (Tenormin; AstraZeneca, England) for heart rates more than 65 beats per minute or 50 mg of atenolol for heart rates less than 65 beats per minute but more than 50 beats per minute 1 hour before CT imaging. Blood pressure, heart rate, and electrocardiogram were monitored, and an additional intravenous dose of 5–30 mg of metoprolol (Lopressor; Novartis, Hampshire, England) was given to achieve a target rate of less than 65 beats per minute. A bolus of 90 mL of iohexol (Omnipaque 350; Amersham Health, Buckinghamshire, England) was injected into an antecubital vein by using an 18-gauge cannula at a flow rate of 6 mL/sec, followed by a flush with 50 mL of saline. A 4-second start delay was used after a threshold value of 160 HU was reached at the root of the ascending aorta. Coronary CT angiography was performed at the first attempt in all but three patients with ectopic beats, and these three cases were rescheduled, with a 2-day course of 100 mg of atenolol once a day.

Quantitative Analysis of Multidetector CT Angiograms
Cross-sectional images were reconstructed with a section thickness of 0.6 mm at 0.4-mm intervals. Retrospective gating was used, and transverse images were reconstructed at 65%, 75%, and 85% of the R-R interval. Additional windows were reconstructed after examination of the data sets if motion artifacts were present. Two sets of CT images were reconstructed with convolution during image data processing. One set was reconstructed with a medium-smooth kernel of B30f and the other with a sharp kernel of B46f. The former images were used to study the distal runoff from the stent, and the latter images were used for the analysis of the in-stent lumen. The best phase of the R-R interval without motion artifact due to irregular beats was used for in-stent lumen evaluation with an off-line independent workstation (Wizard; Siemens). On average, the process took at least an hour for each patient. Intraluminal evaluation of the stent was performed by means of multiplanar reformation of CT data volume. The stent was considered to be occluded if the lumen inside the device appeared darker than the contrast-enhanced vessel lumen proximal to the stent. Nonocclusive in-stent neointimal hyperplasia was indicated by the presence of a darker filling defect between the stent and the contrast-enhanced stent lumen.

Two physicians (K.M.D., W.A.K.D., experienced in cardiac CT for 2.5 and 2.25 years, respectively) independently analyzed all the data sets with multiplanar reformatted images and three-dimensional reformations by using a volume-rendering technique. Both physicians were blinded to the patient's history and to conventional coronary angiographic data, and interobserver variability of the two physician's parameters was calculated by using the coefficient of variation. To improve the delineation of the stent, the images were displayed in a zoom mode at the window level of 400 HU with a window width of 700–1500 HU (8). At each site, measurements were made on magnified transverse images at three adjacent sections, and values were averaged.

The physicians were instructed to look for stent detctability, and, if a stent was detected, to record whether the image of the stent lumen was interpretable or not. Each image of a stent was assigned an image quality score of 1 for excellent quality, 2 for good quality, 3 for moderate quality, or 4 for poor quality. Excellent quality was attributed to the stent appearing as a clear circular or oval area surrounded by low-attenuation fatty tissue without any motion artifact. Good quality was attributed to the presence of discrete blurring of the stent and small streak artifacts emitting shadow on at least one level. Moderate image quality was defined by a blurred stent margin and broader streak artifacts that extended less than 5 mm from the center of the stent. Poor image quality was defined by inadequate delineation between the stent and the surrounding fatty tissue, as well as by streak artifacts that extended at least 5 mm from the center of the stent.

Depending upon the visibility of the stent lumen, restenosis was graded by using a five-point scale as follows: a score of 1 indicated no stenosis; a score of 2 indicated less than 50% stenosis; a score of 3 indicated 50%–74% stenosis; a score of 4 indicated 75% or greater stenosis to total occlusion; and a score of 5 indicated that the stenosis could not be determined. The categorization of the lesion at coronary CT angiography was performed by using subjective assessment of the lumen. A stenosis with a 50% or greater decrease of lumen diameter was considered a significant stenosis. The type of occurrence of neointimal hyperplasia causing stent restenosis was considered focal or marginal; diffuse intrastent involvement within the limit of the stent; diffuse proliferative, extending beyond the stent; or totally occluded (9).

CT attenuation of the in-stent lumen and artery 5 mm proximal and distal to the stent was measured with electronic calipers, and the measurement was correlated with that of the aortic root. The region of interest used for CT attenuation measurements had mean areas of 1 mm² for the in-stent lumen, 3 mm² for the coronary artery lumen, and 5 mm² for the aortic root. For overlapping stents, the distal runoff was measured at the distal stent, and in-stent CT attenuation was measured at each stent. For an interpretable lumen, any in-stent filling defect inside the lumen was mentioned, and average CT attenuation was determined. These defects were examined in both the longitudinal and transverse section of the stent in three consecutive sections, and the values were averaged. In the transverse section, an attempt was made to put the region of interest in the center of the lesion away from the stent strut in a zoomed image. The difference in in-stent CT attenuation of the groups of patent stents and that of stent restenosis was correlated (Table). A receiver operating characteristic curve was plotted for CT attenuation of the poststent artery in the group of stent restenosis to determine a cutoff point. We also looked for any localized nonexpansion of the stent strut. The appearance of the contrast agent in the artery distal to the stent was evaluated visually by using a four-point scale: score of 1, distal artery completely filled with contrast material; 2, moderately filled with contrast material; 3, poorly filled with contrast material; and 4, no appearance of contrast material.

Statistical Analysis
Statistical software (SPSS, version 14.0, 2006; SPSS, Chicago, Ill) was used for the analysis of the data. Means and standard deviations were calculated for continuous variables, whereas frequencies and percentages were calculated for categoric variables. To determine a significant difference among mean levels of CT attenuation at different locations, the Student t test or Mann-Whitney U test, wherever applicable, was performed. ² Tests were performed to determine the association between conventional coronary angiographic data and other categoric variables. Sensitivity, specificity, positive predictive value, and negative predictive value of coronary CT angiography were calculated by using conventional coronary angiography as a reference standard. To rule out clustering in data, a general estimating equation was performed. A statistical package (Stata, version 8.0, 2003; Stata, College Station, Tex) was used for analysis. A P value of less than .05 was considered to indicate a significant difference.

	RESULTS

TOP ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS DISCUSSION ADVANCES IN KNOWLEDGE IMPLICATIONS FOR PATIENT CARE... References

 
Stents
The locations of 110 stents (one to six stents per patient; median, 2.0) in 53 patients were in the left anterior descending artery (n = 44), right coronary artery (n = 34), left circumflex artery (n = 26), obtuse marginal artery (n = 4), diagonal artery (n = 1), and ramus artery (n = 1). In 26 arteries, overlapping stents were present.

In total, six types of stents were evaluated. There were 38 Cypher stents (Cordis, Miami, Fla), 38 Taxus stents (Boston Scientific; Natick, Mass), 18 Liberte stents (Boston Scientific), eight Bx Sonic stents (Cordis), six Coroflex stent (B. Braun, Bethlehem, Pa), and two FlexMaster F1 stents (Abbott Laboratories; Abbott Park, Ill). Of these total 110 stents, 76 stents were drug-eluting stents (Cypher, n = 38; Taxus, n = 38), and the other 39 stents were non–drug-eluting stents. The average diameter and length of the stents were 2.7 mm ± 0.24 and 6.8 mm ± 6, respectively.

Restenosis
Coronary CT angiography correctly depicted the status of 97 of 107 (91%) stents (Fig 1). Images of three stents were not interpretable because of motion artifact (n = 2) and metallic artifact (n = 1), in which the lumen of the stent was completely distorted. As per general estimating equation analysis, there was no clustering in the data, and data were independent for other statistical analysis to be performed. In comparison with those at conventional coronary angiography, sensitivity, specificity, positive predictive value, negative predictive value, and accuracy of coronary CT angiography in the detection of stent restenosis was 96.9% (95% confidence interval [CI]: 82.0%, 99.8%), 88.0% (95% CI: 78.0%, 94.0%), 77.5% (95% CI: 61.1%, 86.6%), 98.5% (95% CI: 90.9%, 99.9%), and 91%, respectively. A patent stent was correctly diagnosed in 66 of 107 (62%) stents, and stent restenosis was correctly diagnosed in 31 of 107 (29%) stents. Of the remaining 13 stents, coronary CT angiography had one false-negative result (due to wall calcification). A false-positive result was seen in nine stents with an in-stent hypointense focal defect for which conventional coronary angiographic results were normal (Fig 2).

View larger version (21K): [in this window] [in a new window] [Download PPT slide]   Figure 1: Flow chart of patient study. CCA = conventional coronary angiography, CCTA = coronary CT angiography.

View larger version (118K): [in this window] [in a new window] [Download PPT slide]   Figure 2a: Images in a 55-year-old man. (a) Curved multiplanar reformatted CT image shows stent in proximal part of right coronary artery with linear low-attenuating filling defect (thin arrows) seen to extend from aortic lumen into proximal part of stent. Another low-attenuating filling defect is seen in middle of stent (thick arrow). (b) Transverse section of stent shows in-stent filling defect (arrow) causing less than 50% narrowing of stent at same location of distal arrow in a. Both filling defects are because of neointimal hyperplasia. (c) Corresponding conventional right coronary angiogram shows normal filling of stent (arrow).

View larger version (69K): [in this window] [in a new window] [Download PPT slide]   Figure 2b: Images in a 55-year-old man. (a) Curved multiplanar reformatted CT image shows stent in proximal part of right coronary artery with linear low-attenuating filling defect (thin arrows) seen to extend from aortic lumen into proximal part of stent. Another low-attenuating filling defect is seen in middle of stent (thick arrow). (b) Transverse section of stent shows in-stent filling defect (arrow) causing less than 50% narrowing of stent at same location of distal arrow in a. Both filling defects are because of neointimal hyperplasia. (c) Corresponding conventional right coronary angiogram shows normal filling of stent (arrow).

View larger version (130K): [in this window] [in a new window] [Download PPT slide]   Figure 2c: Images in a 55-year-old man. (a) Curved multiplanar reformatted CT image shows stent in proximal part of right coronary artery with linear low-attenuating filling defect (thin arrows) seen to extend from aortic lumen into proximal part of stent. Another low-attenuating filling defect is seen in middle of stent (thick arrow). (b) Transverse section of stent shows in-stent filling defect (arrow) causing less than 50% narrowing of stent at same location of distal arrow in a. Both filling defects are because of neointimal hyperplasia. (c) Corresponding conventional right coronary angiogram shows normal filling of stent (arrow).

In 31 stenosed stents, neointimal hyperplasia was responsible for restenosis in 29 of the stents. The distribution pattern of the neointimal hyperplasia in those 29 stents was focal in 20 (marginal, 14 stents; focal, six stents), diffuse in five, and totally occluded in four. The marginal type of neointimal proliferation was seen as edge restenosis at coronary CT angiography in 14 of those 29 stents (Fig 3). The other two stent restenoses were caused by stent strut narrowing (Fig 4). In comparison to degree of stent restenosis depicted at conventional coronary angiography, coronary CT angiography correctly depicted 15 of 18 (83%) stent restenoses with less than 50% narrowing, five of five (100%) stent restenoses with 50%–74% narrowing, and nine of nine (100%) stent restenoses with 75% or greater narrowing or total occlusion of the lumen (Figs 5, 6).

View larger version (132K): [in this window] [in a new window] [Download PPT slide]   Figure 3a: Angiograms in a 63-year-old man. (a) Curved multiplanar reformatted CT image shows neointimal hyperplasia causing less than 50% narrowing of distal edge of left circumflex artery stent (thin arrow). Stent lumen is visible with wall calcification (thick arrow). (b) Conventional coronary angiogram reveals less than 50% narrowing (arrow) at distal edge of stent of left circumflex artery.

View larger version (167K): [in this window] [in a new window] [Download PPT slide]   Figure 3b: Angiograms in a 63-year-old man. (a) Curved multiplanar reformatted CT image shows neointimal hyperplasia causing less than 50% narrowing of distal edge of left circumflex artery stent (thin arrow). Stent lumen is visible with wall calcification (thick arrow). (b) Conventional coronary angiogram reveals less than 50% narrowing (arrow) at distal edge of stent of left circumflex artery.

View larger version (43K): [in this window] [in a new window] [Download PPT slide]   Figure 4a: Images in a 58-year-old man. (a) Curved multiplanar reformatted CT image shows two contiguous stents in left anterior descending artery. One wall of proximal stent strut overlaps with that of distal stent with in-stent narrowing (arrow). (b) Conventional angiogram of left anterior descending artery indicates mild narrowing (arrow) of stent lumen.

View larger version (34K): [in this window] [in a new window] [Download PPT slide]   Figure 4b: Images in a 58-year-old man. (a) Curved multiplanar reformatted CT image shows two contiguous stents in left anterior descending artery. One wall of proximal stent strut overlaps with that of distal stent with in-stent narrowing (arrow). (b) Conventional angiogram of left anterior descending artery indicates mild narrowing (arrow) of stent lumen.

View larger version (136K): [in this window] [in a new window] [Download PPT slide]   Figure 5a: Images in a 52-year-old man. (a) Curved multiplanar reformatted CT image shows neointimal hyperplasia causing narrowing (long thick arrow) just proximal to left anterior descending stent with more than 75% restenosis (long thin arrow) of proximal part of stent. Distal runoff (short arrow) is well seen despite tight stenosis. (b) Transverse section of stent shows thin stream of contrast material (arrow) between low-attenuating neointimal hyperplasia causing more than 75% narrowing of in-stent lumen. (c) Conventional coronary angiogram indicates filling of stent with thin stream of contrast material (thin arrow) showing narrowing at proximal artery abutting edge of stent and more than 75% in-stent narrowing of lumen (thick arrow).

View larger version (61K): [in this window] [in a new window] [Download PPT slide]   Figure 5b: Images in a 52-year-old man. (a) Curved multiplanar reformatted CT image shows neointimal hyperplasia causing narrowing (long thick arrow) just proximal to left anterior descending stent with more than 75% restenosis (long thin arrow) of proximal part of stent. Distal runoff (short arrow) is well seen despite tight stenosis. (b) Transverse section of stent shows thin stream of contrast material (arrow) between low-attenuating neointimal hyperplasia causing more than 75% narrowing of in-stent lumen. (c) Conventional coronary angiogram indicates filling of stent with thin stream of contrast material (thin arrow) showing narrowing at proximal artery abutting edge of stent and more than 75% in-stent narrowing of lumen (thick arrow).

View larger version (154K): [in this window] [in a new window] [Download PPT slide]   Figure 5c: Images in a 52-year-old man. (a) Curved multiplanar reformatted CT image shows neointimal hyperplasia causing narrowing (long thick arrow) just proximal to left anterior descending stent with more than 75% restenosis (long thin arrow) of proximal part of stent. Distal runoff (short arrow) is well seen despite tight stenosis. (b) Transverse section of stent shows thin stream of contrast material (arrow) between low-attenuating neointimal hyperplasia causing more than 75% narrowing of in-stent lumen. (c) Conventional coronary angiogram indicates filling of stent with thin stream of contrast material (thin arrow) showing narrowing at proximal artery abutting edge of stent and more than 75% in-stent narrowing of lumen (thick arrow).

View larger version (80K): [in this window] [in a new window] [Download PPT slide]   Figure 6a: Images in a 55-year-old woman. (a) Curved multiplanar reformatted CT image shows three overlapping stents in left anterior descending artery. Proximal stent is patent. Distal two stents are totally occluded by low-attenuating neointimal hyperplasia (thin arrow), with localized stent strut narrowing (thick arrow). (b) Transverse section of stent at location of thin arrow in a shows totally occluded in-stent lumen with low-attenuating neointimal hyperplasia (arrow). (c) Conventional coronary angiogram indicates totally occluded distal stents (thin arrow) in left anterior descending artery. Nearby vessel is diagonal branch of left anterior descending artery (thick arrow) that originates just proximal to occluded stents.

View larger version (62K): [in this window] [in a new window] [Download PPT slide]   Figure 6b: Images in a 55-year-old woman. (a) Curved multiplanar reformatted CT image shows three overlapping stents in left anterior descending artery. Proximal stent is patent. Distal two stents are totally occluded by low-attenuating neointimal hyperplasia (thin arrow), with localized stent strut narrowing (thick arrow). (b) Transverse section of stent at location of thin arrow in a shows totally occluded in-stent lumen with low-attenuating neointimal hyperplasia (arrow). (c) Conventional coronary angiogram indicates totally occluded distal stents (thin arrow) in left anterior descending artery. Nearby vessel is diagonal branch of left anterior descending artery (thick arrow) that originates just proximal to occluded stents.

View larger version (150K): [in this window] [in a new window] [Download PPT slide]   Figure 6c: Images in a 55-year-old woman. (a) Curved multiplanar reformatted CT image shows three overlapping stents in left anterior descending artery. Proximal stent is patent. Distal two stents are totally occluded by low-attenuating neointimal hyperplasia (thin arrow), with localized stent strut narrowing (thick arrow). (b) Transverse section of stent at location of thin arrow in a shows totally occluded in-stent lumen with low-attenuating neointimal hyperplasia (arrow). (c) Conventional coronary angiogram indicates totally occluded distal stents (thin arrow) in left anterior descending artery. Nearby vessel is diagonal branch of left anterior descending artery (thick arrow) that originates just proximal to occluded stents.

CT attenuation measured in the in-stent lumen was higher than that of the coronary artery lumen measured proximal and distal to the stent in both groups. The poststent CT attenuation (Fig 7) of stents with 50% or greater narrowing was higher compared with stents with less than 50% stenosis. It was attributed to the relatively higher CT attenuation of the aorta in these cases. In two of the completely occluded stents, significantly low CT attenuation of –7 and –8 HU was recorded (Fig 6).

View larger version (9K): [in this window] [in a new window] [Download PPT slide]   Figure 7: Line diagram shows relationship of CT attenuation at aortic root and poststent coronary artery to percentage of stenosis. Attenuation is higher in stents with 50% or greater stenosis (solid arrow), and it was because of simultaneous higher attenuation (open arrow) recorded at aortic root in these cases.

	DISCUSSION

TOP ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS DISCUSSION ADVANCES IN KNOWLEDGE IMPLICATIONS FOR PATIENT CARE... References

 
Clinical diagnosis of stent restenosis may be difficult in asymptomatic patients. Recent recommendations for invasive coronary test do not suggest routine follow-up angiography in those patients (11). With a relatively poor detection rate of conventional noninvasive tests, such as exercise stress tests, stress echocardiography, and myocardial scintigraphy (12), coronary CT angiography currently has emerged as one of the promising noninvasive tests because of its combination of unprecedented acquisition speed, spatial resolution, and robustness of use (4,6,8,13,14).

Multidetector 64-section CT enabled stent restenosis to be correctly detected in 97 of 107 (91%) stents in our study, with a correct diagnosis of 66 patent stents and 31 stenosed stents and with sensitivity, specificity, positive predictive value, negative predictive value, and accuracy of 96.9%, 88.0%, 77.5%, 98.5%, and 91%, respectively. Gilard et al (15) reported overall interpretable images of the stent lumen in 126 of 190 (66%) stents with 16-section CT. The detection rate of the interpretable images of the in-stent lumen was dependent upon the size of the stents. In their report, images of 81% of the stent lumens were interpretable when the stent size was more than 3 mm, and images of 51% of the stent lumens were interpretable with stents 3 mm in diameter or larger (15). Sensitivity, specificity, positive predictive value, and negative predictive value were 54%, 100%, 100%, and 94%, respectively, for smaller stents and 86%, 100%, 100%, and 99%, respectively, for stents larger than 3 mm in diameter. Schuijf et al (16) reported that images of 50 of 65 (77%) stents were interpretable by using 16-section CT, with sensitivity and specificity of 78% and 100%, respectively, and better in-stent lumen visibility with stents 3 mm or larger. Our study results showed correct assessment of the stent lumen in 97 of 107 (91%) stents, and we did not see any correlation between in-stent lumen visibility and the size of the stents (P = .104). The average diameter of our stents was 2.7 mm ± 0.24. A relevant observation, however, is the high rate of agreement among various published studies (15,16), including ours, of the high negative predictive value (94%–99%) of a negative finding at coronary CT angiography. This high negative predictive value suggests an important role of noninvasive coronary CT angiography for a reliable detection of a significant stent restenosis in patients with equivocal clinical manifestations and findings, who currently would undergo invasive conventional coronary angiography for exclusion of in-stent restenosis.

A luminal diameter reduction of more than 50% due to neointimal hyperplasia is consistent with hemodynamically significant stent restenosis (17). The accuracy of coronary CT angiography was better for stents with 50% or more reduction of in-stent luminal diameter. However, the accuracy of coronary CT angiography was inferior for stents with less than 50% reduction in luminal diameter, because of inclusion of nine stents with focal defects that had apparently normal findings at conventional coronary angiography. Although early neointimal hyperplasia causing less than 50% restenosis with definite in-stent hypointense focal defects may not change the clinical course of patients, it indicates higher sensitivity of the scanner in detecting such early pathologic findings. Until recently, the assessment of the stent lumen for nonocclusive stent restenosis due to neointimal hyperplasia was a challenging task. With the improved isotropic resolution of the present generation of CT, more of these early defects may be detected and an early surveillance may be initiated before progression into a significant lesion. Angiography has limited accuracy in tortuous, overlapping, bifurcational, and eccentric lesions (18). This same phenomenon may explain relatively poor correlation of conventional coronary angiography in early lesions in nine stents with focal in-stent defects detected with CT. Moreover, similar small in-stent filling defects may have been further masked by the attenuation of contrast agent on the double-projection planar image at conventional coronary angiography.

The development of neointimal hyperplasia, which is mainly responsible for stent restenosis, ranges from less than 10% with a drug-eluting stent to 40% with an uncoated metallic stent (19–21). Stent restenosis is largely attributed to thrombus formation and acute inflammation early after stent deployment and with subsequent neointimal hyperplasia (22). Recently, the identification of low-attenuation focal in-stent filling defects at CT due to neointimal hyperplasia has been considered the cause of stent restenosis (23). We have documented similar hypoattenuating in-stent focal defects in 29 stents. Presumably, two of our totally occluded stents with significantly lower CT attenuation (–7 HU and –8 HU) may have been produced by infiltration of fatty plaque. The presence of similar atherosclerotic plaque within a stent lumen is an unusual finding. This could be explained by penetration of struts into the necrotic core, beneath already torn subintima during balloon dilation and by prolapse of plaque material between the stent wires (22). Edge restenosis was the dominant type of stent restenosis in our series, and it was presumed to be due to various factors, such as reduction in local drug availability, deficient lesion covering, and procedure-related trauma or damage to the coating of the stent by calcification or an overlapping stent (23).

We believe that diagnosis of stent restenosis depends on the visualization of the stent strut and the in-stent neointimal hyperplasia. Distal runoff cannot be considered a reliable indicator of patency, especially in nonocclusive stent restenosis. Unlike at conventional coronary angiography, with the venous injection at coronary CT angiography, a considerable amount of retrograde filling of the coronary artery distal to the stent via collateral arteries may obscure the real pathologic lesion inside the stent. Moreover, arrhythmia, inadequate breath holding, and reduced ejection fraction with prolonged transit time may further affect the distal runoff with insufficient contrast enhancement of the poststent artery. Although CT attenuation in the stenosed stents was significantly lower compared with that of patent stents (P < .001), the receiver operating characteristic curve did not reveal a characteristic pattern for a specific cutoff point. This could be because of prolonged transit time and variation of CT attenuation measured at the root of aorta (483 HU ± 71), despite use of contrast material with the same iodine strength and a fixed trigger point of 160 HU with an additional delay of 4 seconds (24) for optimum contrast enhancement. The same variation may be responsible for the relatively higher CT attenuation of the poststent coronary artery in the group of stents with 50%–74% stenosis.

Our study had limitations. First, our gantry speed was 370 msec and isotropic resolution was 0.4 x 0.4 x 0.6 mm. With the option of isotropic resolution of 0.4 x 0.4 x 0.4 mm and gantry speed of 330 msec, which is available in the latest generation of 64-section CT scanners, a better quality image could have been achieved. Second, we did not use integrated scanner software for densitometric evaluations in the time-attenuation analysis of the contrast-enhanced coronary arteries proximal and distal to the stents and the aortic root. Presumably, this may not affect the final analysis of our data because CT attenuation of the distal coronary artery may not be consistent with the grades of stent restenosis due to various factors described earlier in this article. Third, the time of coronary CT angiography and conventional coronary angiography was up to 30 days. During this time, development of stent restenosis was a possibility. However, these patients were not symptomatic at the time of coronary CT angiography, and additional conventional coronary angiography was not performed. Last, we did not have intravenous ultrasonographic correlation of those stents with in-stent filling defects with a normal finding at conventional coronary angiography.

In conclusion, 64-section coronary CT angiography can depict in-stent low-attenuation filling defects with a higher degree of accuracy (91%) and negative predictive value (98.5%; 95% CI: 90.9%, 99.9%), which suggests an important role of noninvasive coronary CT angiography for ruling out significant stent restenosis in patients with otherwise equivocal test findings.

	ADVANCES IN KNOWLEDGE

TOP ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS DISCUSSION ADVANCES IN KNOWLEDGE IMPLICATIONS FOR PATIENT CARE... References

 

• Sixty-four–section coronary CT angiography allowed detection of in-stent neointimal hyperplasia in 29 of 31 restenosed stents, with sensitivity, specificity, positive predictive value, negative predictive value, and accuracy of 96.9%, 88.0%, 77.5%, 98.5%, and 91%, respectively.
  
• In comparison with the degree of stent restenosis at conventional coronary angiography, coronary CT angiography correctly depicted 15 of 18 (83%) stent restenoses with less than 50% narrowing, five of five (100%) stent restenoses with 50%–74% narrowing, and nine of nine (100%) stent restenoses with 75% or greater narrowing or total occlusion of the lumen.
  
• The diagnosis of stent restenosis depends on the visualization of the stent strut and in-stent neointimal hyperplasia; distal runoff cannot be considered a reliable indicator of patency, especially in nonocclusive stent restenosis.
  

	IMPLICATIONS FOR PATIENT CARE

TOP ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS DISCUSSION ADVANCES IN KNOWLEDGE IMPLICATIONS FOR PATIENT CARE... References

 

• Contrast-enhanced 64-section coronary CT angiography can be used as a promising, accurate noninvasive test to evaluate stent restenosis.
  
• We believe, with a negative predictive value of 98.5% and accuracy of 91%, coronary CT angiography can have substantial clinical implications for screening patients suspected of having coronary artery stent restenosis.
  

	FOOTNOTES

 

Abbreviations: CI = confidence interval

Author contributions: Guarantor of integrity of entire study, K.M.D.; study concepts/study design or data acquisition or data analysis/interpretation, all authors; manuscript drafting or manuscript revision for important intellectual content, all authors; manuscript final version approval, all authors; literature research, A.A.E., A.M.S., W.A.K.D., J.A.S.; clinical studies, K.M.D., A.A.E., W.A.K.D., H.A.A.; statistical analysis, A.A.E., R.S., W.A.K.D.; and manuscript editing, A.A.E., W.A.K.D., J.A.S.

Authors stated no financial relationship to disclose.

	References

TOP ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS DISCUSSION ADVANCES IN KNOWLEDGE IMPLICATIONS FOR PATIENT CARE... References

 

1. Ruygrok PN, Webster MW, Valk VD, et al. Clinical and angiographic factors associated with asymptomatic restenosis after percutaneous coronary intervention. Circulation 2001;104:2289–2294. [Abstract/Free Full Text]
2. Regar E, Serruys PW, Bode C, et al. Angiographic findings of the multicenter randomized study with the sirolimus-eluting Bx velocity balloon-expandable stent (RAVEL): sirolimus-eluting stents inhibit restenosis irrespective of the vessel size. Circulation 2002;106(15):1949–1956. [Abstract/Free Full Text]
3. Grube E, Silber S, Hauptmann KE, et al. TAXUS I: six- and twelve-month results from a randomized, double-blind trial on a slow-release polymer-based paclitaxel-eluting stent for coronary artery lesions. Circulation 2003;107:38–42. [Abstract/Free Full Text]
4. Kitagawa T, Fujii T, Tomohiro Y, et al. Noninvasive assessment of coronary stents in patients by 16-slice computed tomography. Int J Cardiol 2006;109(2):188–194. [CrossRef][Medline]
5. Mahnken AH, Buecker A, Wildberger JE, et al. Coronary artery stents in multislice computed tomography: in vitro artifact evaluation. Invest Radiol 2004;39:27–33. [CrossRef][Medline]
6. Mahnken AH, Seyfarth T, Flohr T, et al. Flat-panel detector computed tomography for the assessment of coronary artery stents: phantom study in comparison with 16-slice spiral computed tomography. Invest Radiol 2005;40(1):8–13. [Medline]
7. Seifarth H, Ozgun M, Raupach R, et al. Sixty-four– versus 16-slice CT angiography for coronary artery stent assessment: in vitro experience. Invest Radiol 2006;41(1):22–27. [CrossRef][Medline]
8. Hong C, Chrysant GS, Woodard PK, Bae KT. Coronary artery stent patency assessed with in-stent contrast enhancement measured at multi-detector row CT angiography: initial experience. Radiology 2004;233:286–291. [Abstract/Free Full Text]
9. Mehran R, Dangas G, Abizaid AS, et al. Angiographic patterns of in-stent restenosis: classification and implications for long-term outcome. Circulation 1999;100:1872–1878. [Abstract/Free Full Text]
10. The Thrombolysis in Myocardial Infarction (TIMI) trial: phase I findings. TIMI Study Group. N Engl J Med 1985;312:932–936. [Medline]
11. Scanlon PJ, Faxon DP, Audet AM, et al. ACC/AHA guidelines for coronary angiography: a report of the American College of Cardiology/American Heart Association Task Force on practice guidelines (Committee on Coronary Angiography) developed in collaboration with the Society for Cardiac Angiography and Interventions. J Am Coll Cardiol 1999;33:1756–1824. [Free Full Text]
12. Janosi A, Vertes A. Exercise testing and left main coronary artery stenosis: can patients with left main disease be identified? Chest 1991;100:227–229. [Medline]
13. Maintz D, Grude M, Fallenberg EM, Heindel W, Fischbach R. Assessment of coronary arterial stents by multislice-CT angiography. Acta Radiol 2003;44(6):597–603. [Medline]
14. Seifarth H, Raupach R, Schaller S, et al. Assessment of coronary artery stents using 16-slice MDCT angiography: evaluation of a dedicated reconstruction kernel and a noise reduction filter. Eur Radiol 2005;15(4):721–726. [CrossRef][Medline]
15. Gilard M, Cornily JC, Pennec PY, et al. Assessment of coronary artery stents by 16-slice computed tomography. Heart 2006;92(1):58–61. [Abstract/Free Full Text]
16. Schuijf JD, Bax JJ, Salm LP, et al. Noninvasive coronary imaging and assessment of left ventricular function using 16-slice computed tomography. Am J Cardiol 2005;95(5):571–574. [CrossRef][Medline]
17. Cademartiri F, Mollet N, Lemos PA, et al. Usefulness of multislice computed tomographic coronary angiography to assess in-stent restenosis. Am J Cardiol 2005;96:799–802. [CrossRef][Medline]
18. Mottram PM, Meredith IT. Intravascular ultrasound assessment of ambiguous coronary lesions. Heart Lung Circ 2001;10:58–62. [Medline]
19. Fischman DL, Leon MB, Baim DS, et al. A randomized comparison of coronary-stent placement and balloon angioplasty in the treatment of coronary artery disease. Stent Restenosis Study Investigators. N Engl J Med 1994;331:496–501.
20. Kiemeneij F, Serruys PW, Macaya C, et al. Continued benefit of coronary stenting versus balloon angioplasty: 5-year clinical follow-up of Benestent-I trial. J Am Coll Cardiol 2001;37:1598–1603. [Abstract/Free Full Text]
21. Morice MC, Serruys PW, Sousa JE, et al. A randomized comparison of a sirolimus-eluting stent with a standard stent for coronary revascularization. N Engl J Med 2002;346:1773–1780. [Abstract/Free Full Text]
22. Farb A, Sangiorgi G, Carter AJ, et al. Pathology of acute and chronic coronary stenting in humans. Circulation 1999;99(1):44–52. [Abstract/Free Full Text]
23. Pugliese F, Cademartiri F, Mieghem CV, et al. Multidetector CT for visualization of coronary stents. RadioGraphics 2006;26:887–904. [Abstract/Free Full Text]
24. Fleischmann D. Use of high concentration of contrast media: principles and rationale-vascular district. Eur J Radiol 2003;45(suppl 1):S88–S93. [CrossRef][Medline]

This article has been cited by other articles:

				R. J. Gibbons, P. A. Araoz, and E. E. Williamson The year in cardiac imaging. J. Am. Coll. Cardiol., January 6, 2009; 53(1): 54 - 70. [Full Text] [PDF]

				H. B. Lim, G. Hur, S. Y. Kim, Y. H. Kim, S. U. Kwon, W. R. Lee, and S. J. Cha Coronary Stent Fracture: Detection with 64-Section Multidetector CT Angiography in Patients and in Vitro Radiology, December 1, 2008; 249(3): 810 - 819. [Abstract] [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]

This Article Abstract Figures Only Full Text (PDF) All Versions of this Article: 2452061389v1 245/2/424    most recent Submit a response Alert me when this article is cited Alert me when eLetters are posted Alert me if a correction is posted Citation Map Services Email this article to a friend Similar articles in this journal Similar articles in PubMed Alert me to new issues of the journal Download to citation manager Citing Articles Citing Articles via HighWire Citing Articles via Google Scholar Google Scholar Articles by Das, K. M. Articles by Al Suwaidi, J. Search for Related Content PubMed PubMed Citation Articles by Das, K. M. Articles by Al Suwaidi, J.