Non-invasive computed tomography angiography in the assessment of coronary stent patency: An Australian experience

Background: This study aimed to evaluate the feasibility and accuracy of 16-slice computed tomography (CT) in the assessment of coronary stent patency. CT coronary angiography (CA) has a high degree of accuracy in the assessment of coronary artery disease compared with invasive selective CA. However, its accuracy in the evaluation of stent patency is not well investigated. Methods: We conducted a retrospective observational study of paired CT coronary angiography (CT-CA) and invasive fluoroscopic coronary angiography (FCA) in 37 patients with 47 coronary stents. CT-CA was carried out with an electrocardiogram-gated 16-slice CT (LightSpeed-16, General Electric (GE), WI, USA). Two CT reporters, blinded to the FCA findings, assessed CT images for stent patency. A cardiologist blinded to CT findings reported FCA. FCA was regarded as the reference standard. Results: A CT-CA could assess 45 of 47 coronary stents (96%). Non-assessable stents on CT-CA were due to motion artefacts and stent-blooming effects. Of those 45 assessable stents, CT-CA correctly identified five out of seven stents with binary in-stent restenosis (ISR) and 37 of 38 stents without binary ISR. The sensitivity and specificity of 16-slice CT in the evaluation of coronary stents for binary ISR were 71% (95% confidence interval (CI) (29%, 96%)) and 97% (95%CI (86%, 100%)), respectively, exclusive of non-assessable stents. The positive and negative predictive values of 16-slice CT were 83% (95%CI (36%, 100%)) and 95% (95%CI (83%, 99%)), respectively. Conclusion: Sixteen-slice CT has a low sensitivity, but very a high specificity when compared with FCA in the evaluation of coronary stents for ISR.