Diagnostic accuracy of 64-slice CT in the assessment of coronary stents

Purpose. The purpose of this study was to assess the diagnostic accuracy of 64-slice computed tomography (64-CT) coronary angiography in the detection of coronary in-stent restenosis. Materials and methods. Ninety-five patients (72 men and 23 women, mean age 58 8 years) with previous percutaneous coronary intervention with stenting and suspected restenosis underwent 64-CT (Sensation 64, Siemens). The mean time between stent deployment and 64-CT was 6.1 +/- 4.2 months. The scan parameters were: slices 32x2, individual detector width 0.6 mm, rotation time 0.33 s, feed 3.84 mm/rotation, 120 kV, 900 mAs. After the intravenous administration of iodinated contrast material (Iomeprol 400 mgI/ml, Iomeron, Bracco) and a bolus chaser (40 ml of saline), the scan was completed in < 12 s. All coronary segments with a stent were assessed on 64-CT by two observers in consensus and judged as: patent, with intimal hyperplasia (lumen reduction of < 50%), with in-stent restenosis (>= 50%), or with in-stent occlusion (100%). The consensus reading was compared with conventional coronary angiography. Results. Four patients were excluded because of insufficient image quality. In the remaining 91, we assessed 102 stents (31 RCA; 10 LM; 54 LAD; 7 CX). In 14 (13.7%) stents, in-stent restenosis (n=8) or in-stent occlusion (n=6) was found. Intimal hyperplasia was detected in 11 (10.8%) stents. The sensitivity and negative predictive value of 64-CT for in-stent occlusion were 100% and 100%, respectively, whereas for all stenoses, > 50% they were 92.9% and 98.7%, respectively.