Assessment of in-stent restenosis using 64-MDCT: analysis of the CORE-64 Multicenter International Trial

Abstract

Objective: Evaluations of stents by MDCT from studies performed at single centers have yielded variable results with a high proportion of unassessable stents. The purpose of this study was to evaluate the accuracy of 64-MDCT angiography (MDCTA) in identifying in-stent restenosis in a multicenter trial.

Materials and methods: The Coronary Evaluation Using Multidetector Spiral Computed Tomography Angiography Using 64 Detectors (CORE-64) Multicenter Trial and Registry evaluated the accuracy of 64-MDCTA in assessing 405 patients referred for coronary angiography. A total of 75 stents in 52 patients were assessed: 48 of 75 stents (64%) in 36 of 52 patients (69%) could be evaluated. The prevalence of in-stent restenosis by quantitative coronary angiography (QCA) in this subgroup was 23% (17/75). Eighty percent of the stents were <or=3.0 mm in diameter.

Results: The overall sensitivity, specificity, positive predictive value, and negative predictive value to detect 50% in-stent stenosis visually using MDCT compared with QCA was 33.3%, 91.7%, 57.1%, and 80.5%, respectively, with an overall accuracy of 77.1% for the 48 assessable stents. The ability to evaluate stents on MDCTA varied by stent type: Thick-strut stents such as Bx Velocity were assessable in 50% of the cases; Cypher, 62.5% of the cases; and thinner-strut stents such as Taxus, 75% of the cases. We performed quantitative assessment of in-stent contrast attenuation in Hounsfield units and correlated that value with the quantitative percentage of stenosis by QCA. The correlation coefficient between the average attenuation decrease and >or=50% stenosis by QCA was 0.25 (p=0.073). Quantitative assessment failed to improve the accuracy of MDCT over qualitative assessment.

Conclusion: The results of our study showed that 64-MDCT has poor ability to detect in-stent restenosis in small-diameter stents. Evaluability and negative predictive value were better in large-diameter stents. Thus, 64-MDCT may be appropriate for stent assessment in only selected patients.

Fig. 1. 42-year-old symptomatic man with known coronary artery disease who had bare metal 3.0 × 20.0 mm stent (Lekton, Biotronik, GmbH) placed 8 months before MDCT angiography and quantitative coronary angiography (QCA). Image quality is good

A, MDCT angiography image used for qualitative assessment shows 50% restenosis, B, Quantitative assessment of centerline plot at 2-mm intervals throughout segment and stent shows metal markers at stent edges (arrowheads) with ballooning artifacts preventing evaluation of inflow and outflow. Attenuation in stent shows drop from 350 HU (short arrows) to 225 HU (long arrow), indicating in-stent restenosis of 50%. C, Stenosis was assessed on QCA image to be 61%.

Fig. 2. Correlation of attenuation within stents with invasive quantitative coronary angiography (QCA)

A, Correlation of QCA and minimum CT attenuation value in Hounsfield units within stents (p = 0.091). Correlation coefficient is −0.2419. Regression line: Y = minimum attenuation in Hounsfield units × (−0.0637445) + 52.47177. B, Correlation of QCA and average CT attenuation value in Hounsfield units within stents (p = 0.073). Correlation coefficient is −0.2555. Regression line: Y = average attenuation in Hounsfield units × (−0.089731) + 72.1787.

Fig. 3. 62-year-old symptomatic man with coronary artery disease who had bare metal 3 × 15 mm stent placed in left circumflex coronary artery 6 months 23 days before studies

A, MDCT angiography image quality was good. Qualitative and quantitative assessments show restenosis of 50% and 27%, respectively. B, Quantitative coronary angiography assessment shows 73% in-stent restenosis (arrows).

Fig. 4. 73-year-old symptomatic man with coronary artery disease who had drug-eluting 2.5 × 24 mm stent (Taxus, Boston Scientific) placed in left circumflex coronary artery 6 months and 10 days before studies. Image quality was good

A, Qualitative and quantitative assessments by MDCT angiography did not show in-stent restenosis. B, Quantitative coronary angiography shows no in-stent restenosis (arrows). Inset shows area of interest magnified.