Assessment of coronary artery bypass grafts: value of different breath-hold MR imaging techniques

Our aim was to evaluate the patency of coronary artery bypass grafts and to detect graft stenosis using different breath-hold MR imaging techniques.

Twenty-two patients with 59 grafts (14 internal mammary artery grafts and 45 saphenous vein grafts) and 76 distal anastomoses (singular and sequential grafts) were studied using a 1.5-T scanner. A two-dimensional T2-weighted breath-hold half-Fourier acquisition single-shot turbo spin echo (HASTE) sequence and a three-dimensional breath-hold contrast-enhanced MR angiography sequence (fast imaging with steady-state free precession) were performed. For MR angiography, a bolus of 20 ml of gadopentetate dimeglumine was used. Time delay for contrast injection was calculated by a test bolus. The gold standard was coronary angiography performed within 14 days of MR imaging. All images were evaluated independently by two radiologists.

With the HASTE sequence, 95% of the patent grafts were recognized (42/44); specificity was 93% (14/15). MR angiography had both a sensitivity (41/44) and specificity (14/15) of 93%. Interobserver agreement for both sequences was good (Cohen's kappa = 87%; McNemar test, p = 56%). Forty-nine (83%) of 59 patent distal graft anastomoses were revealed with the HASTE sequence; 38 (64%) of 59 were seen on contrast-enhanced angiography. With HASTE imaging, only two of eight hemodynamically significant graft stenoses were detected. MR angiography revealed only four of eight significant graft stenoses.

The HASTE sequence and three-dimensional MR angiography proved to be useful MR techniques when evaluating the patency of coronary artery bypass grafts. However, reliable detection of graft stenosis does not yet seem possible with these imaging techniques.