Abstract 6106: Assessment of In-Stent Re-Stenosis Using Mdct 64 in Patients from the Core64 Multicenter International Study
Background: Evaluation of stents by MDCT in single-center trials has yielded variable results with many studies showing high proportion of un-evaluable stents due to metallic and motion artefacts. The purpose of this study was to evaluate the accuracy of 64-slice CT scanner in identifying in-stent restenosis in “all-comers” in a multicenter trial.
Methods and Results: The CorE-64 Multicenter Trial evaluated the accuracy of 64-detector row MDCT in 291 patients clinically referred for conventional coronary angiography for the assessment of coronary artery disease. A total of 75 stents in 52 patients were assessed. 36 out of 52 patients (69%) and 48 out of 75 stents (64%) could be adequately evaluated by MDCT. The prevalence of in-stent restenosis bby quantitative coronary angiography was 31% (15/48) in this population. Overall sensitivity, specificity, PPV and NPV to detect 50% in stent stenosis by MDCT as compared to QCA by invasive angiography was: 36.4%, 71.1%, 26.7% and 72.4% with overall accuracy of 63.3% for the 48 evaluable stents. 80% of the stents evaluated were ≪ 26>3.0 mm in diameter. Ability to evaluate the stent varied across stent sizes (60% for stents ≪26>3.0 and 67% for stents >3.0) and stent types: thick strut stents such as BX Velocity 50%, Cypher 62.5%, and thinner strut Taxus 75%. Per patient analysis revealed: a sensitivity of 33% and specificity of 87.5%. The sensitivity and specificity varied from 22–44% and from 71–77% between the two readers, respectively. In the attempt to increase accuracy of the qualitative visual assessment we performed detailed quantitative assessment of in-stent contrast attenuation measured in Hounsfield units (HU) and correlated this with quantitative percent stenosis by angiography. Correlation coefficient between average HU decrease and >50% stenosis by QCA was only – 0.25 (p=0.073). Thus quantitative assessment failed to improve accuracy of MDCT over qualitative assessment by two experienced independent readers.
Conclusions: In this multicenter trial 64-slice CT appears to have limitations in assessing in-stent restenosis. The use of this technology cannot be recommended for routine stent assessment until further improvements are made in imaging quality.