The Clinical Application Value Of320-Row Computed Tomography Coronary Angiography To Assess In-Stent Restenosis

Objective The study sought to evaluate the diagnostic performance of320-row CT coronaryangiography (CTA) in the evaluation of significant in-stent restenosis (≥50%luminal narrowing)referred for quantitative coronary angiography (CAG). Methods We studied69patients withprevious stent implantation who underwent both CTA and CAG. We assessed diagnosticperformance for in-stent restenosis (ISR) with CTA in comparison with CAG as the standard ofreference using quantitative approach. In addition, to assess the CTA stent image quality usinggrade, to evaluate the relationship of the CTA stent image quality and the diagnostic coincidencerate with the stent characteristics, heart rate and the style of stent implantation. Results There were110stents for evaluation. CAG identified14stents with significant in-stent restenosis. CTAcorrectly identified13stents with significant in-stent restenosis, underestimated the degree ofrestenosis in1stent and incorrectly diagnosed5stents with significant in-stent restenosis. Besides,6stents were of nondiagnostic image quality. Accordingly, including6stents of nondiagnosticimage quality, sensitivity, specificity, positive and negative predictive value were93%,89%,54%and99%, respectively. Excluding6stents of nondiagnostic image quality, sensitivity, specificity,positive and negative predictive value were93%,94%,72%and99%, respectively. Thepercentages of good and moderate CTA stent image quality (56%vs.27%,25%vs.49%) anddiagnostic coincidence rate（95%vs.78%）both were significant different between stent diameter≥3.0mm and <3.0mm(both P＜0.05). The percentages of poor CTA stent image quality (12%vs.45%) and diagnostic coincidence rate（94%vs.76%）both were significant different betweenstent strut thickness＜140μm and≥140μm (both P＜0.05). The percentages of good, moderateand poor CTA stent image quality (53%vs.40%,26%vs.39%,21%vs.21%) and diagnosticcoincidence rate (91%vs.88%) both were not significant different between heart rate≥65beats/min and <65beats/min (both P＞0.05). The percentages of poor CTA stent image quality(36%vs.17%) were significant different between overlapping and bifurcations stents and single stent (P＜0.05), and the diagnostic coincidence rate (86%vs.90%) was not associated with thestyle of stent implantation (P＞0.05). Conclusions Our results indicate320-row CTA allowsaccurate noninvasive assessment of significant in-stent restenosis in selected patients. Stents with alarge diameter and thin struts made image quality and the diagnostic performance better.