Bypass Graft and Native Postanastomotic Coronary Artery Patency: Assessment With Computed Tomography

Background Multidetector computed tomography has been shown to be useful in the evaluation of coronary artery bypass grafts in previous studies. We studied the accuracy of multidetector computed tomography in the detection of patency and significant stenosis of both grafts and native postanastomotic coronary arteries. Methods Ninety-six patients with 216 grafts (98 left mammary artery, 8 right mammary artery, 8 radial artery, and 102 venous grafts) were investigated by 16-slice computed tomography. Native postanastomotic coronary arteries were also evaluated. Patients unable to maintain a breath hold of 40 s were excluded. Computed tomography data were compared with the results of conventional angiography. Results On a segment-based model, the overall feasibility of computed tomography was 98.1% (212 of 216 grafts) for bypass grafts and 93.1% (201 of 216 segments) for postanastomotic coronary arteries. The leading cause of unfeasibility for postanastomotic coronary arteries was the small diameter of the examined vessel (<1.5 mm). Computed tomography correctly diagnosed all the 25 occluded grafts. Of the 33 significant stenoses of grafts, computed tomography correctly diagnosed 31. Sensitivity, specificity, positive predictive value, and negative predictive value were 100%, 98.5%, 96.5%, and 100%, respectively, for bypass graft; and 100%, 97.7%, 85%, and 100%, respectively, for coronary arteries. On a patient-based model, the feasibility, sensitivity, specificity, positive predictive value, and negative predictive value were 89.4% (86 of 96 patients), 100%, 93%, 86.4%, and 100%, respectively. Conclusions Multidetector computed tomography allows a very accurate assessment of arterial and venous conduits and of postanastomotic native coronary arteries in patients with previous bypass graft. Despite high feasibility (93.1%), limitations of the method were breath-hold duration (35 to 40 s) and postanastomotic assessment of small vessels (which, however, precluded the analysis in only 4.6% of cases).