Breath-hold FLASH and FISP cardiovascular MR imaging: left ventricular volume differences and reproducibility

To compare fast imaging with steady-state precession (FISP) and fast low-angle shot (FLASH) magnetic resonance acquisitions to quantify left ventricular volumes, mass, and function and to determine if the two techniques are comparable.

Left ventricular volume studies were performed in 10 patients with heart failure and in 10 healthy subjects by using FISP and FLASH imaging. Identical section positions were used for section-by-section contour comparisons. Manual analysis was performed by two experienced observers. The study was repeated on a different day and interobserver and interstudy reproducibility assessed.

With FISP, end-diastolic volume was larger (healthy subjects: +18 mL [13%], P <.001; patients: +6 mL [3%], not significant), end-systolic volume larger (healthy subjects: +9 mL [17%], P =.001; patients: +8 mL [6%], P =.001) and left ventricular mass smaller (healthy subjects: -25 g (19%), P <.001; patients: -21 g (11%), P <.001). There were no significant differences in ejection fraction. Both sequences had excellent interstudy and interobserver reproducibility, with statistically better reproducibility for interstudy healthy-subject ejection fraction on FISP images (P =.05). Section-by-section analysis determined that at FISP, endocardial contours were drawn larger and the epicardial contours smaller than on FLASH images. FISP enabled better delineation of epicardial fat from myocardium, of blood-myocardium interface in areas of trabeculation or papillary muscles, and of the atrioventricular ring.

FISP produces small but significantly higher left ventricular volume measurements, as compared with FLASH imaging. FLASH imaging and FISP have similar reproducibility.