Revascularization in patients with stable angina is often guided by either myocardial-perfusion cardiovascular MRI or invasive angiography to measure fractional flow reserve (FFR). The unblinded, multicentre MR-INFORM trial was designed to compare these two approaches. A total of 918 patients with typical angina and cardiovascular risk factors were randomly assigned to cardiovascular MRI or measurement of FFR. In the cardiovascular MRI group, 40.5% of patients met the criteria to undergo revascularization (ischaemia in ≥6% of the myocardium), whereas 45.9% of patients in the FFR group met the revascularization criteria (FFR ≤0.8). Fewer patients in the cardiovascular MRI group than in the FFR group underwent index revascularization (35.7% versus 45.0%; P = 0.005). The primary composite outcome (death, nonfatal myocardial infarction or target-vessel revascularization within 1 year) occurred in 3.6% of patients in the cardiovascular MRI group and 3.7% of patients in the FFR group, which met the threshold for noninferiority of cardiovascular MRI compared with FFR.