[Magnetic resonance angiography. Value for evaluating the patency of aortocoronary bypass]

Assessment of coronary bypass patency with conventional angiography can be limited by difficulties in selective catheterism. The aim of this work was to determine whether ultrarapid sequences proposed for imaging coronary arteries could be used to evaluate bypass patency.

Twenty patients (mean age 66 +/- 7 years) with 42 aorto-coronary bypasses (32 venous and 10 mammary bypasses) underwent conventional angiography and magnetic resonance angiography within a 1 week interval. All patients were evaluated 1 to 15 years after bypass surgery. Magnetic resonance imaging (MRI) was performed in ventral supine position using a flexible surface coil. Breathhold electrocardiogram-synchronized sequences were acquired using a 1.5 T magnet, with TR = 156 ms, TE = 8 ms, slice thickness 5 mm, 126.256 matrix, and a 7 to 11 segmentation per cardiac cycle. At most, 24 transverse slices were acquired for each patient.

Twenty-six patent bypasses were correctly identified with MRI. Three bypass were patent on angiography and appeared occluded on MRI (2 mammary and 1 venous bypass). Nine bypasses were classed as occluded with both techniques. MRI demonstrated patency in one bypass for which selective catheterism was not possible. The sensitivity of MRI for the diagnosis of patency in aorto-coronary bypasses was 89% with a 92% specificity. MRI did not identify 1 out of 4 bypass stenoses.

We conclude that the ultrarapid breath-hold magnetic resonance sequences can be used to determine patency of aorto-coronary bypasses with good sensitivity and specificity. Our results would suggest however that this method is inadequate for the diagnosis of bypass stenosis.