Skip to main content

Advertisement

Free-breathing myocardial T1 mapping using magnetization-prepared slice interleaved spoiled gradient echo imaging

Background

Quantitative myocardial T1 mapping and extracellular volume fraction (ECV) show promise for non-invasive assessment of cardiomyopathies. Most available T1 mapping sequences use a single slice breath-hold acquisition with balanced steady state free precession (b-SSFP) readout [1]. However, b-SSFP readout is sensitive to B0 field inhomogeneity and is potentially T2 dependent [1]. In this study, we sought to investigate the feasibility of a free breathing multi-slice T1 mapping sequence using slice-interleaved spoiled gradient echo (GRE) imaging.

Methods

The proposed sequence used multiple inversion recovery (IR) experiments. In each IR experiment, a non-selective inversion pulse is applied and followed by the acquisition of 5 slices over the next 5 heart beats, and 3 rest cycles [2]. This IR experiment is repeated 5 times using different slice orders to obtain signal samples at TI, TI + 1 RR, TI + 2 RR, TI + 3 RR, TI + 4 RR. This block of 5 IR experiments is finally repeated using a different TI value. The fully recovered longitudinal magnetization is also initially acquired for each slice without any IR pulse (∞ image). Respiratory motion was corrected using prospective slice tracking and retrospective image registration. ECG-triggered single shot acquisitions were used with GRE readout (TR/TE/α=4.3/2.1ms/10˚, FOV=280×272 mm2, voxel size=2×2 mm2, slice thickness=8 mm, 5 slices, 43 phase-encoding lines, linear ordering, 10 linear ramp-up pulses, SENSE factor=2.5, half Fourier=0.75, bandwidth=382Hz/pixel). For comparison, MOLLI [3] was acquired with a b-SSFP readout and similar parameters (except TR/TE/α=2.6/1.3ms/70°, 1 slice, bandwidth=1785 Hz/pixel). Imaging was performed on a 1.5 T Philips scanner. T1 accuracy, precision, and reproducibility were evaluated in simulations and phantom. In-vivo spatial variability and reproducibility of native T1 mapping was measured in 11 healthy adult subjects (35±21y, 4 m), imaged 5 times with each sequence. Three of these subjects were also imaged at ~15min after contrast injection to demonstrate the feasibility of ECV mapping.

Results

The proposed sequence provided improved accuracy and similar precision than MOLLI in both simulation and phantom experiments (accuracy: p=0.01; precision: p=0.16). MOLLI was more reproducible in phantom (p<0.001). In-vivo, the proposed sequence yielded higher native T1 times than MOLLI (1094±24ms vs. 1010±27ms, p<0.001) with similar spatial variability (58±7ms vs. 61±9ms, p=0.44) and reproducibility (25±9ms vs. 17±8ms, p=0.15). ECV measurements were 0.21±0.01 using the proposed sequence.

Conclusions

Free breathing multi-slice T1 mapping using a magnetization-prepared slice interleaved spoiled GRE imaging is feasible and yields similar in-vivo precision/reproducibility as MOLLI but with improved accuracy. In addition, the proposed sequence allows simultaneous imaging of 5 slices within free-breathing in 100 sec.

Figure 1
figure1

Accuracy, precision and reproducibility of the proposed sequence obtained in Monte Carlo simulation (20,000 repetitions, fixed T2 of 50 ms, SNR corresponding to 50 in the ∞ image) (a) and phantom experiments (set of vials with NiCl2 doped agarose, 15 repetitions of the sequence) (b). Results were compared to the MOLLI (5-(3)-3 scheme) sequence. Accuracy was measured in each vial as the difference between spin echo T1 measurements and the average T1 over all 15 repetitions. Precision was measured in each vial as the average (over all 15 repetitions) of the standard deviation of T1 within a vial. Reproducibility was measured in each vial as the standard deviation (over all 15 repetitions) of the mean T1 within a vial. Improved accuracy and similar precision were achieved using the proposed sequence in both simulations and phantom experiments. T1 mapping reproducibility was slightly decreased with the proposed sequence.

Figure 2
figure2

In-vivo native T1 and ECV mapping using the proposed sequence. Example of multi-slice T1 maps and ECV maps obtained in one healthy subject is shown in (a). Homogeneous T1 map quality was achieved in all slices for all five repetitions. Homogeneous ECV map quality was also observed through all slices. Native T1 measurements, spatial variability, and reproducibility obtained using the proposed sequence, are reported in average over all subjects in (b). Each metric was quantified using a 16 myocardial segment model in all subjects by analysis of the three mid-ventricular slices. Spatial variability was measured for each segment as the average (over the five repetitions of all subjects) of the standard deviation of T1 measurements within that segment. Reproducibility was measured for each segment as the average (over all subjects) of the standard deviation (over the 5 repetitions) of the mean T1 time of that segment.

References

  1. 1.

    Kelmann : JCMR. 2014

  2. 2.

    Weingärtner : MRM. 2014

  3. 3.

    Messroghli : MRM. 2004

Download references

Author information

Correspondence to Sébastien Roujol.

Rights and permissions

This article is published under license to BioMed Central Ltd. This is an Open Access article distributed under the terms of the Creative Commons Attribution License (http://creativecommons.org/licenses/by/4.0), which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited. The Creative Commons Public Domain Dedication waiver (http://creativecommons.org/publicdomain/zero/1.0/) applies to the data made available in this article, unless otherwise stated.

Reprints and Permissions

About this article

Verify currency and authenticity via CrossMark

Keywords

  • Balance Steady State Free Precession
  • Extracellular Volume Fraction
  • Inversion Recovery Pulse
  • Single Shot Acquisition
  • Inversion Recovery Experiment