Added value of rest to stress study for recognition of artifacts in perfusion cardiovascular magnetic resonance.

BACKGROUND The objective was to determine whether rest perfusion (RP) adds to stress perfusion (SP) and late gadolinium enhancement (LGE) cardiac magnetic resonance (CMR) for detection of impaired coronary flow reserve. METHODS We enrolled patients (n=45) referred for myocardial perfusion SPECT (MPS) for adenosine CMR stress. SP, RP and LGE images were obtained with 99mTc sestamibi injection during a single adenosine infusion. Segmental perfusion and confidence scores were recorded for SP-LGE interpreted with and without RP. CMR agreement with MPS was determined. RESULTS MPS was normal in 653 and abnormal in 67 segments. SP-LGE CMR interpreted without RP was normal in 407, abnormal in 313 segments, and showed poor agreement with MPS (58%). Two hundred thirty-seven segments were changed to normal using data from RP, improving agreement (87%, p<0.0001). Reader confidence was low in 33 patients with SP-LGE and improved in 26 patients using SP-RP-LGE, where 37/45 were read with high confidence. Artifact was present in 68% of SP CMR and accounted for false positive studies. CONCLUSION Agreement between single stress adenosine CMR and MPS is optimized by combining RP, LGE and SP CMR. Addition of RP CMR to SP-LGE CMR improved agreement with MPS and reader confidence. Improved CMR pulse sequences may change the role of rest perfusion data.