A four-point ordinal-scale qualitative flow index was used for assessment of patency of 75 coronary artery bypass grafts in 26 patients examined with spiral computed tomography (CT). CT findings were compared with selective graft angiographic findings. Of 54 open grafts, 52 were patent at initial selective graft an-giography and 50 were patent at spiral CT; accuracy rates were 97% (73 of 75) and 95% (71 of 75), respectively. Spiral CT flow index agreed with angiographically determined flow in 85% (95% Cl: 0.77, 0.93) of grafts. The K statistic demonstrated very good to excellent intermodality (0.75) and interobserver (0.89) agreement. Spiral CT may be a feasible means of assessing quality of flow in bypass grafts.
The use of computed tomography (CT) in the assessment of coronary artery bypass graft (CABG) patency was discussed more than 20 years ago (1). Subsequently, the visualization of CABGs with CT by using either sequential imaging of the heart or dynamic CT has been widely reported, with varying success and a sensitivity range of 45%-100% and with contrast material volumes of 25-150 ml (2-5). Findings in many studies indicated increased sensitivity as the number of levels imaged was increased (6-8). In particular, since the left anterior descending
artery graft often arches upward over the main pulmonary artery, it may be missed if an appropriate level is not chosen. Thus, this missing of grafts at imaging contributes to the false-positive rate observed with CT (9). Although it should follow that there would be greater sensitivity for any potential volumetric imaging technique, this has generally been precluded by the large amounts of contrast agent that would be required with use of previous CT technology. In each study, CT results were directly correlated with angiographic assessment.
Spiral CT involves continuous scanning while the patient is simultaneously advanced through the gantry during a single breath hold. Acquired data can be reconstructed in the axial plane at varying section positions at any point in the scanning cycle, with no apparent difference in spatial resolution between conventional and volumetric scans (10). As the intravenously administered contrast material travels through the graft segments, good image detail is achieved, and this achievement of detail allows visualization of entire coronary artery graft segments during one injection. Earlier work has demonstrated the comparable accuracy of spiral CT and selective graft angiography in establishing patency rates (11,12). The purpose of this article was to report our experience and technique with spiral CT in the assessment of a qualitative coronary artery bypass graft flow index and the purpose of this study was to determine its accuracy compared with that of selective graft angiography.