[The value of nuclear magnetic resonance tomography in evaluating the patency of aortocoronary bypass grafts].

BACKGROUND

Patients with previous coronary artery bypass graft surgery often present with pain of unknown origin. Noninvasive documentation of graft patency is of obvious clinical importance.

METHODS

In order to assess the efficacy of magnetic resonance imaging (MRI) in evaluating graft patency, 44 patient with prior coronary artery bypass graft surgery and a history of chest pain were studied by coronarography and MRI. MRI was performed within 2.1 +/- 3.8 days from coronarography using a 0.5 Tesla magnet, Spin-Echo T1 technique, cardiac and respiratory gating and scannings in transaxial plans. A graft was defined as patent, if a signal void was identified in at least two different slices in a position consistent with a bypass graft. Images were analyzed by two different observers aware of the type of surgery but not the result of the coronarography.

RESULTS

Eighty-nine out of 100 grafts were classified correctly by MRI. Sixty-three grafts were patent as shown by coronarography; 59 of them were classified correctly by MRI. Thirty-seven grafts were shown as occluded; 30 of them were classified correctly by MRI. In particular, 43 out of 45 grafts to the left anterior descending artery, 18 of them using the internal mammary artery, 23 out of 30 graft to the left circumflex artery and 23 out 25 grafts to the right coronary artery were classified correctly.

CONCLUSIONS

This study demonstrates the capability of MRI to evaluate coronary artery bypass graft patency with a sensitivity of 94% and specificity of 81%; this technique has significantly clinical limitations because resolution is not adequate to evaluate the presence of graft stenosis.