We compared contrast-enhanced MRI (CeMRI) with the most widely used imaging techniques for myocardial infarct (MI) diagnosis, SPECT and Echo, in unselected patients with chronic coronary artery disease (CAD). Two blinded operators assessed scars on MRI, SPECT and Echo images using a 16-segments LV model. We studied 105 consecutive patients: 50 had Q-wave MI (Q-MI), 19 non Q-wave MI or rest angina (nonQ-MI/RA) and 36 effort angina (EA) history. CeMRI was positive, respectively, in 96%, 37%, and 6%, SPECT in 90%, 53%, and 44%, and Echo in 84%, 32%, and 28% of patients (within Q-MI: CeMRI vs. SPECT p < 0.03, vs. Echo p < 0.001; within EA CeMRI vs. SPECT and ECHO p < 0.001; all trends p < 0.001, pseudo r-square: 0.56-0.75 for CeMRI, 0.18-0.28 for SPECT and 0.23-0.37 for Echo). CeMRI and SPECT agreed in 83 patients (79%); negative SPECT with 1 +/- 0 segments subendocardial delayed enhancement (DE) was found in 4 (4%); negative CeMRI with 4 +/- 3 segments perfusion defects in 18 (17%), 16 of whom were obese or showed LBB or sub-occlusion of related coronary. CeMRI and Echo agreed in 78 patients (75%); negative Echo with 2 +/- 1 segments subendocardial DE was found in 13 (12%) and negative CeMRI with 11 +/- 7 segments kinetic abnormalities in 14 (13%), in 10 confirmed by Cine-MRI. In Q-MI, CeMRI detects DE more frequently than perfusion defects and, especially, kinetic abnormalities are found by SPECT and Echo, respectively. CeMRI identifies small areas of DE also in some patients with nonQ-MI or RA but usually not in patients with EA. This biologically plausible decreasing trend is shown by CeMRI more clearly than by SPECT and Echo. Disagreement between CeMRI and SPECT or Echo may be reduced, but perhaps not fully eluded, performing dobutamine Echo and SPECT after maximal epicardial coronary dilatation.