Contemporary Discrepancies of Stenosis Assessment by Computed Tomography and Invasive Coronary Angiography

Abstract

Background Ongoing advancements of coronary computed tomographic angiography (CTA) continue to challenge the role of invasive coronary angiography (ICA) as the gold standard for the evaluation of coronary artery disease (CAD). We sought to investigate the diagnostic accuracy of 320-slice CTA for detecting obstructive CAD in reference to ICA and nuclear myocardial perfusion imaging using single-photon emission computed tomography. Methods For the CORE320 study (Coronary Artery Evaluation Using 320-Row Multidetector Computed Tomography Angiography and Myocardial Perfusion), 381 patients at 16 centers underwent CTA, nuclear myocardial perfusion imaging by single-photon emission computed tomography, and ICA for the evaluation of CAD. Imaging studies were analyzed in blinded core laboratories, and a stenosis of ≥50% by quantitative coronary angiography was considered obstructive, whereas a stress difference score of ≥1 indicated inducible myocardial ischemia. The area under the receiver operating characteristic curve was used to evaluate diagnostic accuracy. Results Of 381 patients, 229 (60%) had obstructive CAD by quantitative coronary angiography. Diagnostic accuracy of CTA on a per-patient analysis revealed an area under the receiver operating characteristic curve of 0.90 (95% CI, 0.87-0.93). Per-vessel and per-segment analysis revealed lower area under the receiver operating characteristic curve of 0.87 (0.84-0.90) and 0.81 (0.78-0.83), respectively. Median radiation dose was lower for CTA versus ICA: 3.16 (interquartile range, 2.82-3.59) versus 11.97 (interquartile range, 7.60-17.8) mSv ( P<0.001). Accuracy for identifying patients with inducible myocardial ischemia by SPECT-MPI was similar for CTA and ICA (area under the receiver operating characteristic curve, 0.68 versus 0.71 by quantitative coronary angiography and 0.68 by visual angiographic assessment; P>0.05). Furthermore, accuracy for identifying patients who subsequently underwent clinically driven coronary revascularization also was similar for CTA (0.76 [0.71-0.81]) and ICA (0.78 [0.74-0.83]; P=0.20). Conclusions Contemporary CTA accurately identifies patients with obstructive CAD by ICA at lower radiation exposure; however, agreement is lower in vessel- and segment-level analyses. Both CTA and ICA perform similarly for predicting clinically driven revascularization and for detecting myocardial ischemia by myocardial perfusion imaging using single-photon emission computed tomography, suggesting that limitations by both CTA and ICA contribute to variability of stenosis quantification. Clinical Trial Registration URL: https://www.clinicaltrials.gov . Unique identifier: NCT00934037.

Keywords: ROC curve; computed tomography angiography; coronary angiography; coronary artery disease; humans.