Abstract
Background: Multidetector computed tomography angiography has emerged as a rapidly developing method for the noninvasive detection of coronary artery disease.
Purpose: To investigate the diagnostic accuracy and limitations of multidetector computed tomography (MDCT) in the detection of significantly (>50%) obstructive coronary artery disease (CAD) using a scanner equipped with 16×0.625-mm collimation.
Material and Methods: MDCT angiography was performed in 153 patients (99 male, 54 female; mean age 55±10 years) with suspected CAD and scheduled for conventional coronary angiography (CCA). Image quality was assessed in terms of artifacts and segment visibility, and the assessable segments were screened for the presence of significant stenoses (>50% lumen diameter reduction). The diagnostic performance of MDCT for the detection of significant stenosis was compared with the results of CCA.
Results: In all 153 patients, MDCT was carried out without complications. A total of 1989 coronary artery segments were evaluated. After exclusion of 394 non-evaluable segments (19.8%), 1595 segments (80.2%) were included in the analysis. The most frequent causes of poorly assessable segments were motion artifact (36%) and severe calcification (23%). Considering only the segments judged to be evaluable, the sensitivity, specificity, and positive and negative predictive values of 16-slice MDCT were 85%, 97%, 79%, and 98%, respectively. Including all segments in the analysis (evaluable and non-evaluable), the sensitivity was 74%, specificity 96%, positive predictive value 73%, and negative predictive value 97%.
Conclusion: When all coronary artery segments are included, 16-slice MDCT has moderate sensitivity and very high specificity and negative predictive value in assessing coronary artery stenoses. High specificity and negative predictive value indicate that 16-slice MDCT may be a useful tool in reliably ruling out significant lesions in patients with a low pretest probability.