Objective: Our goal was to validate cardiac measurements derived from multishot echo planar MRI (EPI) as compared with the well validated conventional GRE technique.
Method: Ten healthy subjects underwent breath-hold EPI and non-breath-hold GRE imaging in the short axis orientation of the left ventricle (LV) on a standard 1.5 T MR system. Ten section levels were obtained to encompass the entire LV. Measurements were obtained of end-diastolic volume (EDV), end-systolic volume (ESV), stroke volume (SV), ejection fraction (EF), LV mass, time to end-systole (TES), and peak filling rate (PFR). Analysis of variance was performed to determine agreement between GRE- and EPI-derived measurements.
Results: The acquisition time for EPI was much shorter than that for GRE (2.5 vs. 15 min). Both imaging techniques yielded good quality images allowing LV volumetrics. Agreement between GRE and EPI was best for measurements of EDV, SV, and LV mass; somewhat less agreement was found for ESV, EF, TES, and PFR. The intraobserver variability for measuring TES and PFR was higher for GRE than EPI (one sided F test; critical values at p = 0.05 were > 3.18).
Conclusion: Multishot EPI of the heart provides accurate measurements of LV function and mass in a time-efficient manner.