Synopsis

Conventionally, evaluation of myocardial T₁-times is limited to a single snapshot of the cardiac cycle, leaving much of the dependence between functional and tissue characterization unstudied. Here, we propose an ECG-triggered steady-state Look-Locker technique that allows for functional, cardiac phase-resolved native T₁-mapping. Integratedly acquired phase-resolved B₁⁺-maps are used for T₁-time correction. High accuracy and good consistency of the T₁-times across cardiac phases is shown in phantom scans. In-vivo T₁-times show slight underestimation and similar precision compared to saturation-recovery T₁-mapping. High visual image quality at all cardiac phases is obtained at temporal resolutions up to 40ms in a single breath-hold.

Background

Methods

Sequence: In the proposed Look-Locker CINE (LLCine) sequence, T₁-quantification is based on magnetization readout during the approach to pulsed steady-state. Initially, magnetization is driven to steady-state using FLASH excitations, and an inversion pulse is applied (Fig. 1). Then imaging pulses are played continuously, while the pulsed inversion-recovery curve is sampled with segmented k-space sampling for each cardiac phase. After reaching steady-state again, magnetization is re-inverted, and the next k-space segment is acquired. This recovery to pulsed steady-state spans several heartbeats. Hence, multiple time points on the inversion-recovery curve are sampled, allowing for spatially- and temporally-resolved quantification of T₁-time throughout cardiac cycle. Dummy pulses without signal acquisition fill the end of cardiac cycles after acquiring the last cardiac phase, to maintain pulsed steady-state, and minimize effects of R-R interval variability. Reconstruction: Apparent relaxation during continuous FLASH excitations is described⁶ by T₁^*

$$M_z(t)=M_0\left(1–B\cdot e^{-\frac{t}{T_1^*}}\right)$$ $$B=1-\cos(\alpha_{inv})$$ $$\frac{1}{T_1^*} = \frac{1}{T_1} - \frac{\log( \cos( \alpha ) ) }{T_R}$$

Hence, actual T₁ can be derived from this sequence, given TR and excitation flip angle (FA). Efficiency of the rectangular inversion pulse (B in Eq. 2 above) can be used to estimate B₁⁺-inhomogeneities and calculate the actual FA as:

$$\frac{\alpha}{\alpha^{nom}}\approx\frac{\alpha_{inv}}{\alpha_{inv}^{nom}}$$ $$T_1=\left(\frac{1}{T_1}-\frac{\log(\cos(\frac{\arccos(1-B)}{\alpha_{inv}^{nom}}\cdot\alpha_{nom}))}{T_R}\right)^{-1}$$

Numerical Simulations: The conditioning of extracting T₁ from T₁^* depends on TR and FA. Numerical simulations were performed to minimize T₁-error sensitivity by optimizing these two parameters. Bloch-simulations for the proposed sequence were performed for TR=3.5,4, …15.0ms and FA=1°,2°,...20°. T₁-times were then calculated from 1000 noisy simulations, and evaluated for accuracy and precision at all parameter values. Imaging: All imaging was performed at 3T (Siemens Magnetom Prisma) and the following imaging parameters were used in the remainder of the study: TR/TE/FA=5/2.5ms/3°, FOV=300x300mm², resolution=1.9x1.9mm², slice thickness=10mm, temporal resolution=40-60ms, breath-hold duration=19-23s. SAPPHIRE T₁-mapping⁷ was performed for comparison. Phantom imaging was performed to test T₁-mapping accuracy, and consistency of the T₁ estimate across the cardiac phases. In-vivo imaging of a mid-ventricular short-axis slice was performed on 10 healthy subjects (4male, 32±15y/o). T₁-times were evaluated using septal ROIs drawn for each cardiac phase. In-vivo precision was defined as T₁-time variation within the ROIs.

Results

Numerical simulations indicate diminished precision and accuracy for short TR and high FA, both causing rapid apparent relaxation. T₁-precision is also compromised with low FA due to insufficient signal. Optimal T₁-sensitivity, at breath-hold scan-times was achieved at TR/FA=5ms/3°.

In phantom experiments (Fig. 2), LLCine is in good agreement with SAPPHIRE T₁-mapping (NRMSE=3.1%) and highly consistent across phases (Coefficient-of-Variation (CoV)=2.1±0.9%). In-vivo T₁-maps display a high level of homogeneity in the myocardium and good delineation towards the blood pool at all phases (Fig. 3). In-vivo T₁-times after B₁⁺-correction fall into the expected range at 3T, while uncorrected T₁-maps exhibit underestimation (Fig. 4). Minor variations of the in-vivo T₁-times are observed throughout the cardiac cycle (CoV=1.9%). In-vivo precision is comparable to SAPPHIRE T₁-mapping, with slightly higher variability at late diastolic phases. B₁⁺-maps generated by the proposed approach are visually homogeneous and largely T₁-insensitive, with increased variability towards the end of cardiac cycle.

Discussion and Conclusion

This study demonstrates the feasibility of cardiac phase-resolved T₁-mapping at temporal resolutions as fine as 40ms in a single breath-hold. The integrated phase-resolved B₁⁺-maps were leveraged to correct for flip-angle deviations, in order to maximize accuracy.

The sequence parameters were numerically optimized for the expected T₁ range at 3T. Hence, if tissue with T₁-times other than native myocardium, e.g. post-contrast, is to be evaluated, a dedicated parameter optimization should be performed to allow for adequate recovery durations and accuracy.

In the proposed sequence, sample-points on the inversion-recovery curve are dependent on cardiac phase. Later cardiac phases lack acquisition of short inversion times, hence increased in-vivo variability was observed. However, since myocardium remains largely stationary during diastolic quiescence, temporal acquisition/reconstruction window can be increased in later phases, for improving SNR to mitigate this effect.

Acknowledgements

References

1. Jellis CL, Kwon DH. "Myocardial T1 mapping: modalities and clinical applications." Cardiovasc Diagn Ther. 2014 Apr;4(2):126-37.

2. Higgins DM, Moon JC. "Review of T1 Mapping Methods: Comparative Effectiveness Including Reproducibility Issues." Curr Cardiovasc Imaging Rep (2014) 7: 9252.

3. Ferreira VM, Wijesurendra RS, Liu A, Greiser A, Casadei B, Robson MD, Neubauer S, Piechnik SK. "Systolic ShMOLLI myocardial T1-mapping for improved robustness to partial-volume effects and applications in tachyarrhythmias." J Cardiovasc Magn Reson. 2015 Aug 28;17:77.

4. Reiter U, Reiter G, Dorr K, Greiser A, Maderthaner R, Fuchsjäger M. "Normal diastolic and systolic myocardial T1 values at 1.5-T MR imaging: correlations and blood normalization." Radiology. 2014 May;271(2):365-72.

5. Hamlin SA, Henry TS, Little BP, Lerakis S, Stillman AE. "Mapping the future of cardiac MR imaging: case-based review of T1 and T2 mapping techniques." Radiographics. 2014 Oct;34(6):1594-611.

6. Kaptein R, Dijkstra K, Tarr CE. “A single-scan Fourier transform method for measuring spin-lattice relaxation times.” J Magn Reson 1976; 24:295-300

7. Weingärtner S, Akçakaya M, Basha T, Kissinger KV, Goddu B, Berg S, Manning WJ, Nezafat R. "Combined saturation/inversion recovery sequences for improved evaluation of scar and diffuse fibrosis in patients with arrhythmia or heart rate variability." Magn Reson Med. 2014 Mar;71(3):1024-34.