JCMR 2020;22:62. Prevalence and pattern of myocardial late enhancement in cardiac magnetic resonance of highly trained endurance athlete-s12968-020-0600-1 Sub-segmental

s12968-020-0600-1 Sub-segmental

Pdf Summary

This study tested whether breaking the myocardium into smaller CMR perfusion analysis units improves detection of obstructive coronary artery disease (CAD) compared with the standard 16-segment American Heart Association model.<br /><br />Thirty patients with invasive evidence of obstructive CAD and 20 healthy controls underwent stress perfusion CMR at 3T during adenosine vasodilation. The investigators compared several segmentation schemes: 16 segments, 32 epicardial/endocardial subsegments, 48 circumferential segments, and 96 combined sub-(sub)segments. Perfusion was quantified using Fermi deconvolution, and the diagnostic performance of each method was assessed with ROC analysis.<br /><br />Results showed that finer segmentation substantially improved accuracy. The area under the curve increased from 0.820 for the 16-segment model to 0.937 for 32 segments, 0.973 for 48 segments, and 0.993 for 96 segments. The 96-segment approach also improved the endocardial/epicardial flow ratio, with an AUC of 0.979 versus 0.932 for the 32-segment ratio. The standard 16-segment method had slightly better reproducibility, but the loss in reproducibility was modest and did not outweigh the gain in diagnostic performance.<br /><br />The authors conclude that subsegmental myocardial quantification can better capture the extent and severity of ischemia, improves objective cutoff-based reporting, and may support more reliable databasing and automated analysis. A practical advantage is that this approach can work from a single stress-only perfusion acquisition, reducing contrast dose and scan time. However, the findings are based on a small proof-of-concept study and require validation in larger clinical populations.

Keywords

myocardial perfusion CMR

obstructive coronary artery disease

subsegmental analysis

stress perfusion imaging

adenosine vasodilation

ROC analysis

Fermi deconvolution

endocardial epicardial ratio

diagnostic accuracy

16-segment model