64-Slice Multidetector-row Computed Tomography in the Diagnosis of Coronary Artery Disease Interobserver Agreement Among Radiologists With Varied Levels of Experience on a Per-patient and Per-segment Basis

Purpose: To assess the interobserver variability of 4 radiologists with different levels of experience in the evaluation of 64-slice coronary computed tomographic angiography (cCTA). Materials and Methods: Two board-certified radiologists with 10 and 8 years of experience in reading cCTA and 2 radiology residents, 1 with 3 years of experience in reading cCTA and 1 with experience in reading general computed tomographic scans but without dedicated cCTA training, participated in the study. All the observers independently analyzed 50 cCTA studies for signs of coronary artery disease (stenosis of 0%, <= 49%, 50% to 74%, 75% to 99%, or 100%). Diagnostic accuracy of the 4 readers for stenosis detection on cCTA was compared with that of conventional angiography on a per-segment and per-patient basis. No patients, vessels, or segments were excluded from analysis. Results: On a per-segment basis, correlation between cCTA and invasive coronary angiography was good for readers with more than 10 (r = 0.75), more than 8 (r = 0.75), and more than 3 (r = 0.73) years of cCTA experience. The correlation coefficient was poor (r = 0.39) for the untrained reader. Sensitivity was not significantly (P = 0.56) different between observers with more than 8 and more than 10 years of experience but was significantly (P > 0.05) lower for the reader with less than 3 years experience and for the untrained reader. However, we found no significant difference in overall diagnostic accuracy on a per-patient (P = 0.86) and on a per-segment level (P = 0.72) among the 4 readers. Conclusion: The level of experience significantly influences the sensitivity of coronary artery stenosis detection at cCTA, and thus highlights the need for dedicated training in cCTA interpretation.