Evaluation of Agreement between 64-slice Computed Tomography Angiography and 201-Tl Single Photon Emission Computed Tomography-myocardial Perfusion Imaging in the Diagnosis of Significant Coronary Artery Disease

─ 541 ─ Department of Radiology, Ajou University School of Medicine Department of Nuclear Medicine, Ajou University School of Medicine Department of Cardiology, Ajou University School of Medicine Received February 16, 2011 ; Accepted April 18, 2011 Address reprint requests to : Doo Kyoung Kang, M.D., Department of Diagnostic Radiology, Ajou University School of Medicine, San 5, Woncheon-dong, Yeongtong-gu, Suwon 442-721, Korea. Tel. 82-31-219-5849 Fax. 82-31-219-5862 E-mail: kdklsm@ajou.ac.kr Purpose: To compare coronary computed tomography angiography (CTA) with single photon emission computed tomography-myocardial perfusion imaging (SPECT-MPI) for the detection of physiologically significant coronary artery disease (CAD). Materials and Methods: We evaluated 202 patients undergoing 64-slice coronary CTA and 201-Tl SPECT-MPI within a 3-month time interval. In addition, 68 patients underwent invasive coronary angiography (ICA). Coronary artery stenoses with luminal narrowing ≥50% were defined as “significant” on CTA and ICA. All myocardial segments were classified as reversible or fixed perfusion defects and normal segments on 201-Tl SPECT-MPI, and were allocated to the corresponding coronary vessels. Agreement and diagnostic performance between each imaging modality for physiologically significant CAD was calculated using the kappa (κ) statistic and receiver operating characteristic analysis, respectively. Results: The sensitivity and specificity of CTA for the detection of physiologically significant CAD were 88% and 86% by patient-based analysis, and 84% and 91% by vessel-based analysis as compared to 201-Tl SPECT-MPI, respectively. The agreement between CTA and SPECTMPI was good (κ= 0.647) and moderate (κ= 0.558) by patientand vessel-based analyses, respectively. The accuracy of CTA for predicting perfusion defects on SPECT-MPI was comparable (area under the curve; 0.814 vs. 0.819, p=0.902 on patient-based analysis, and 0.808 vs. 0.749, p=0.197 on vessel-based analysis) to ICA. Conclusion: Coronary stenosis ≥50% on coronary CTA shows good agreement with perfusion defects in SPECT-MPI.