Coronary magnetic resonance angiography (CMRA) continues to make broad advances but is still not ready for routine clinical usage, according to a new study published online in the Journal of the American College of Cardiology.
The new study reports on the single-center experience with a new 3.0 tesla contrast MRA in 96 patients scheduled for conventional angiography. The procedure was successfully performed in 62 patients. CMRA correctly identified CAD in 32 patients and ruled it out in 23 patients. Here are the results on a per-patient basis:
- sensitivity= 94.1% (32 of 34)
- specificity= 82.1% (23 of 28)
- accuracy= 88.7% (55 of 62)
Commenting on the study for CardioBrief, PK Shah agreed that the study showed “incremental improvement in the detection of coronary stenosis” but “it is not a home run” because only 64% of eligible patients could be imaged. “That is a lot of drop outs,” said Shah.
Shah also objected to the use of 50% as a cut off for binary stenosis and recommended a 70% cut off instead. “Although overall negative predictive accuracy for detecting a >50% stenosis was relatively high per segment (98%), it was 92% on a per patient basis, meaning that 8% of patients with greater than 50% stensois would be missed, which is not great.”
“While it is better than prior MRI studies it is still no cigar either as a reliable screening test or for clinical decision making,” he concluded. “More work is needed.”
Speak Your Mind