Usefulness of 64-detector computed tomographic angiography for diagnosing in-stent restenosis in native coronary arteries.

The purpose of this study was to evaluate the accuracy of detector computed tomographic angiographic qualitative and quantitative analyses for the detection of in-stent restenosis (ISR) Previous studies have used qualitative analyses exclusively and have excluded "unevaluable" stents. Multidetector computed tomographic angiography (MDCT) was performed before quantitative coronary angiography in 67 patients with 132 stents that were evaluated by 2 techniques: (1) qualitative, on the basis of degree of visual hypodensity, and (2) quantitative, comparing in-stent with prestent Hounsfield units. All stents were evaluated, irrespective of image quality. The incidence of ISR was 12.5%. The sensitivity (94%), specificity (74%), and positive predictive value (39%) of the qualitative evaluation were superior to the quantitative technique (82%, 54%, and 21%, respectively); negative predictive values were similar (99% vs 95%). Accuracies were equal in stents located in proximal and distal vessels. In conclusion, ISR can be evaluated qualitatively by 64-slice MDCT with excellent sensitivity and negative predictive accuracy without exclusion of unevaluable stents and with reasonable specificity but low positive predictive value. Quantitative analysis was less accurate.