Heart-rate independent myocardial T1-mapping using combined saturation and inversion preparation pulses

Sebastian Weingärtner, Mehmet Akçakaya, Sophie Berg, Kraig V. Kissinger, Warren J. Manning, Reza NezafatJournal of Cardiovascular Magnetic Resonance 2013 · DOI

Abstract

Accurate myocardial T1-mapping is essential for characterizing heart tissue, yet standard techniques like MOLLI are highly sensitive to heart-rate variations. This paper introduces SAPPHIRE (SAturation Pulse Prepared Heart rate independent Inversion-REcovery), a novel MRI sequence that utilizes combined saturation and inversion pulses to eliminate magnetization history. By "resetting" the tissue state at each heartbeat, SAPPHIRE achieves T1 measurements that are invariant to heart rate, offering a more robust diagnostic tool for cardiac assessment.

The Problem: Heart-Rate Bias in T1-Mapping

In cardiac MRI, T1-mapping allows clinicians to identify myocardial fibrosis or edema. The current standard, Modified Look-Locker Inversion Recovery (MOLLI), uses an inversion recovery scheme to sample magnetization. However, MOLLI assumes the tissue has sufficient time to fully recover between preparations. When a patient's heart rate increases, this recovery time is truncated, leading to a residual "magnetization history" that biases the resulting T1 values. This heart-rate dependency can lead to inconsistent data, particularly in pre-contrast (native) T1 scans.

The SAPPHIRE Solution

The researchers developed the SAPPHIRE sequence to act as a physical "reset" for the heart's magnetization. The key innovation is the integration of a saturation pulse immediately following the R-wave of the ECG cycle.

The SAPPHIRE sequence diagram shows how saturation (Sat.) and inversion (Inv.) pulses are timed with the ECG to reset magnetization history before data sampling.

As illustrated in the sequence diagram, the saturation pulse dephases the magnetization to a known zero state, effectively erasing any influence from previous heartbeats. This is followed by an inversion pulse and a single-shot image readout. To capture a wider range of inversion times ($T_{inv}$), some data points are sampled in the heart cycle immediately following the preparation. Additionally, a baseline cycle without preparation provides a reference to improve the mathematical T1 fit.

Key Results and Validation

The sequence was tested against MOLLI using both phantom models and healthy human volunteers.

Phantom Stability

In a bottle phantom with a known T1 of ~1300 ms, researchers simulated heart rates from 60 to 120 beats per minute (bpm).

T1 measurements for MOLLI (red) and SAPPHIRE (blue) across simulated heart rates. MOLLI values drop significantly as heart rate increases, while SAPPHIRE remains stable.

The results demonstrated that while MOLLI's T1 values dropped linearly as the heart rate increased ($r = -0.99$), SAPPHIRE's measurements remained remarkably stable ($r = -0.10$). At 60 bpm, both methods were within 2% of each other, but at 120 bpm, MOLLI's bias became severe, whereas SAPPHIRE accurately maintained the phantom's true T1 value.

In Vivo Performance

In studies with five healthy volunteers, SAPPHIRE produced T1-maps that were qualitatively more homogeneous than those from MOLLI. Numerically, there was no statistically significant difference between the T1 values of the two methods at resting heart rates ($p > 0.10$), suggesting that SAPPHIRE is a viable clinical alternative that provides greater reliability during heart-rate fluctuations.

Conclusion and Trade-offs

SAPPHIRE successfully decouples T1-mapping from heart-rate physiology, providing a "heart-rate independent" metric for cardiac tissue characterization. The primary trade-off is an increased acquisition time; SAPPHIRE requires breath-holds of 16–23 seconds, compared to 12–17 seconds for MOLLI. Despite this longer duration, the elimination of heart-rate-induced bias represents a significant advancement for consistent and accurate cardiac diagnostics.

Summary updated

Abstract

Figure 1 Sequence diagram depicting the SAPPHIRE T1-mapping sequence: a saturation pulse is performed after the R-wave to erase the magnetization history.It is followed by the inversion pulse and a single-shot image readout.To extend the range of applicable inversion times the data readout of some SAPPHIRE experiments is performed in the heart-cycle after the magnetization preparation.Additionally the first heartcycle is performed without magnetization preparation and the last heart-cycle with the saturation pulse only.This increases the effective inversion times and improves the T1 fit.
BibTeX
@article{weingartner2013heart,
  author = {Sebastian Weingärtner and Mehmet Akçakaya and Sophie Berg and Kraig V. Kissinger and Warren J. Manning and Reza Nezafat},
  title = {{Heart-rate independent myocardial T1-mapping using combined saturation and inversion preparation pulses}},
  journal = {Journal of Cardiovascular Magnetic Resonance},
  year = {2013},
  doi = {10.1186/1532-429x-15-s1-p46},
}
Figure