Fully Refocused Gradient Recalled Echo (FRGRE): Factors Affecting Flow and Motion Sensitivity in Cardiac MRI

Abstract

The Fully Refocused Gradient Recalled Echo (FRGRE) magnetic resonance pulse sequence, previously reported as True FISP, was implemented and reported on a GE Signa CV/i 1.5T scanner. The purpose of this research was to optimize the pulse sequence design and scanning parameters to improve image quality for cardiac applications. A 2-D, multi-slice, multi-phase, breathold, segmented k-space, prospectively gated FRGRE sequence was implemented with TE and TR values as short as 0.9 and 3.0 msec, respectively. Pulse sequence design changes were investigated theoretically in terms of moment calculations and by assessing the quality of long- and short-axis cardiac images. The most influential scanning parameter for the improvement of image quality was both a short TR and TE. Placement of the z gradient rephaser immediately before the slice select gradient reduced the accumulated moments and improved image quality. Increasing the slice thickness from 3 to 8-10 mm significantly reduced flow artifact. The number of views per segment was doubled, without decreasing image quality, cutting breathold time in half compared to cardiac scanning with conventional sequences with longer TRs. The optimal flip angle was approximately 60 degrees. The use of a full or fractional echo had no noticeable effect on image quality. Surprisingly, the addition of flow compensation pulses significantly decreased image quality. In summary, optimization of sequence design and scanning parameters significantly reduced flow artifacts seen in FRGRE images.

Quantification of aortic valve stenosis in MRI—comparison of steady-state free precession and fast low-angle shot sequences

“…Primarily, they criticise the use of an echo time of 4.8 ms which results in so-called spin dephasing artefacts which display the transstenotic jet as a (gray to black) signal void which makes it difficult to reliably measure the aortic valve area. Moreover, Friedrich et al recommend the use of a steady-state free precession (SSFP) sequence with shorter echo times which is less susceptible to this problem [14] and thus more easily identifies the orifices. In a recent study, they have shown that MR imaging using a SSFP sequence reliably visualizes the area of the valvular orifice in symptomatic patients with aortic valve stenosis [15].…”

Feasibility of Cardiovascular Magnetic Resonance to Assess the Orifice Area of Aortic Bioprostheses

“…Nevertheless, SSFP is generally the preferred sequence for assessing bioprostheses because of its superior signal-to-noise ratio, clear-cut bloodtissue contrast, and high spatial and temporal resolution, making the accurate identification of the fast-moving delicate valve cusps easier. 28,29 Schlosser et al 30 demonstrated that the assessment of aortic valve stenoses using SSFP correlates better with TEE compared with FGRE. However, when flow artifacts render SSFP images of nondiagnostic quality, we showed that the application of FGRE can improve image quality and allow image analysis in select cases.…”

Extrinsic multiecho phase-contrast SSFP: evaluation on cardiac output measurements

“…Another option would be to expand to a three-echo sequence and determine the velocity map as Nielsen and Nayak [23] did; however, this would require an even longer TR. The sensitivity to artifacts from through-plane flow might be reduced using a second bipolar gradient after the second readout to make the sequence flow compensated again [12,13]. The multiecho PC-SSFP sequence, as was described in this paper, might be better suitable for cerebrospinal fluid flow measurements in the brain, as velocities are lower, pulsatility is less and uniform surrounding tissue is present for phase offset measurement.…”

“…The uncompensated bipolar gradient and the long TR made the sequence sensitive to fast flowing and pulsatile blood [12,13]. To illustrate the effect of the uncompensated gradient, we additionally turned the velocity encoding gradient off.…”

“…A disadvantage of SSFP is its sensitivity to artifacts from pulsatile and high through-plane flow [10][11][12][13]. In the multiecho implementation, this sensitivity is slightly increased due to the longer TR and an uncompensated bipolar gradient.…”