Diagnostic capability of feature-tracking cardiovascular magnetic resonance to detect infarcted segments: a comparison with tagged magnetic resonance and wall thickening analysis.

AIM

To examine the diagnostic capabilities of feature-tracking cardiovascular magnetic resonance (FT-CMR), tagged cine magnetic resonance (MR), and wall thickening (WT) analyses to detect infarcted segments in patients with established myocardial infarction (MI).

MATERIALS AND METHODS

Twenty patients with established MI were selected retrospectively and the peak endocardial circumferential strain (CS) was quantified based on the 16-segment model. According to CMR with late gadolinium enhancement, segments were categorised as transmural MI, subendocardial MI, and no MI.

RESULTS

A total of 320 segments (62 transmural MI, 50 subendocardial MI, and 208 no MI) were analysed. Peak endocardial CS was significantly lower for transmural MI compared with subendocardial MI (p<0.05) and no MI (p<0.001). Cut-off values of -11.2% for CS by FTCMR, -10.9% for CS by tagged MR, and 23.8% for %WT, differentiated between infarcted and non-infarcted segments with a sensitivity of 72%, 71%, and 56%; specificity of 71%, 75%, and 67%; accuracy of 72%, 73%, and 63%; positive predictive value of 57%, 60%, and 48%; negative predictive value of 83%, 83%, and 74%; and an area-under-the-curve of 0.77, 0.79, and 0.64, respectively.

CONCLUSIONS

FT-CMR was diagnostically superior to %WT, and could differentiate between subendocardial and transmural MI. Unlike tagged MR, FT-CMR did not require the acquisition of additional sequences.