[Skip to Content]
Access to paid content on this site is currently suspended due to excessive activity being detected from your IP address Please contact the publisher to request reinstatement.
[Skip to Content Landing]
Original Contribution
May 25, 2005

Noninvasive Coronary Angiography With Multislice Computed Tomography

Author Affiliations

Author Affiliations: Department of Diagnostic Radiology, University Hospital, Ulm, Germany (Drs Hoffmann, Shi, Schmitz, Schmid, Brambs, and Aschoff); and Department of Cardiology, Heart-Center, Ulm (Drs Lieberknecht, Schulze, Ludwig, Kroschel, Jahnke, and Haerer).

JAMA. 2005;293(20):2471-2478. doi:10.1001/jama.293.20.2471

Context Multislice computed tomography (MSCT) has recently evolved as a modality for noninvasive coronary imaging.

Objective To assess the accuracy and robustness of MSCT vs the criterion standard of invasive coronary angiography for detection of obstructive coronary artery disease.

Design, Setting, and Patients Prospective, single-center study conducted in a referral center setting in Germany and enrolling 103 consecutive patients (mean age, 61.5 [SD, 9.7] years) from November 2003–August 2004 who were undergoing both invasive coronary angiography and MSCT using a scanner with 16 detector rows.

Main Outcome Measures Blinded results for both modalities compared using the patient as the primary unit of analysis, with supplementary segment- and vessel-based analyses.

Results One thousand three hundred eighty-four segments (≥1.5 mm diameter) were identified by invasive coronary angiography; nondiagnostic image quality of MSCT was identified for only 88 (6.4%) of these segments, mainly due to faster heart rates. Compared with invasive coronary angiography for detection of significant lesions (>50% stenosis), segment-based sensitivity, specificity, and positive and negative predictive values of MSCT were 95%, 98%, 87%, and 99%, respectively. Quantitative comparison of MSCT and invasive coronary angiography showed good correlation (r = 0.87, P<.001), with MSCT systematically measuring greater-percentage stenoses (bias, +12%). In the patient-based analysis, the area under the receiver operating characteristic curve was 0.97 (95% confidence interval, 0.90-1.00), indicating high discriminative power to identify patients who might be candidates for revascularization (>50% left main artery stenosis and/or >70% stenosis in any other epicardial vessel). Threshold optimization allowed either detection of these patients with 100% sensitivity at a reasonable false-positive rate (specificity, 76.5%; MSCT stenosis, >66%) or optimization of both the sensitivity and specificity (>90%; MSCT stenosis, >76%).

Conclusions Multislice computed tomography provides high accuracy for noninvasive detection of suspected obstructive coronary artery disease. This promising technology has potential to complement diagnostic invasive coronary angiography in routine clinical care.