Feasibility of relaxation along a fictitious field in the 2nd rotating frame (TRAFF2) mapping in the human myocardium at 3 T

João Tourais, Maša Božić‐Iven, Yidong Zhao, Qian Tao, Iain Pierce, Christian Nitsche, George Thornton, Lothar R. Schad, Thomas A. Treibel, Sebastian Weingärtner, Mehmet AkçakayaFrontiers in Cardiovascular Medicine 2024 · DOI

TL;DR

This paper introduces and validates a new MRI technique called TRAFF2 mapping for assessing myocardial tissue properties at 3 Tesla. It demonstrates that TRAFF2 mapping can be performed in a single breath-hold with acceptable image quality and shows promise for detecting myocardial infarction without the need for contrast agents, although further work is needed to mitigate artifacts.

What Problem Does This Paper Address?

Detecting and characterizing myocardial infarction (MI) is crucial for patient management. The current gold standard, late gadolinium enhancement (LGE), relies on injecting contrast agents (gadolinium-based contrast agents, GBCAs). However, GBCAs have limitations, including contraindications for patients with renal insufficiency, potential for nephrogenic systemic fibrosis, and adverse reactions. Furthermore, LGE provides only relative contrast and can be subjective, leading to inter-reader variability.

Researchers have been exploring non-contrast MRI techniques, such as native T1 and T2 mapping, to assess myocardial tissue. However, these methods have limitations in sensitivity and specificity for detecting MI. Another promising avenue is rotating frame relaxation time (T1ρ) mapping, which is sensitive to slower molecular processes. However, T1ρ mapping at high magnetic field strengths (like 3T) is challenging due to strict Specific Absorption Rate (SAR) limits and susceptibility to magnetic field inhomogeneities. This paper addresses the need for a robust, non-contrast MRI method for myocardial tissue characterization at 3T that overcomes the limitations of LGE and existing non-contrast techniques, particularly concerning SAR and field inhomogeneities.

Methods

The core innovation presented is the TRAFF2 (T1 Relaxation Along a Fictitious Field in the 2nd rotating frame) mapping sequence. This technique utilizes a novel pulse sequence design that employs nested amplitude-modulated (AM) and frequency-modulated (FM) RF pulses. These pulses create a "fictitious field" in a second rotating frame, allowing for relaxation measurements with significantly lower SAR requirements compared to traditional T1ρ methods.

The TRAFF2 sequence was designed to be acquired within a single breath-hold (approximately 17 seconds) and involves acquiring five ECG-gated images: one fully relaxed baseline image, three images with varying TRAFF2 preparation durations, and one image after a saturation pulse to model magnetization recovery. Pixel-wise TRAFF2 maps are generated using a three-parameter exponential fitting model.

The study evaluated the TRAFF2 sequence in three phases:

  1. Phantom studies: To assess accuracy, precision, and intra-scanner repeatability against reference TRAFF2, T1, and T2 measurements. The robustness to B0 (static magnetic field) and B1+ (radiofrequency field) inhomogeneities was also tested.
  2. Healthy subjects: To evaluate in vivo feasibility, image quality, precision, reproducibility, and inter-subject variability across three short-axis slices.
  3. Patients with suspected cardiovascular disease: To assess clinical feasibility by comparing TRAFF2 maps with LGE, native T1, T2, and ECV mapping in a small cohort.

Datasets

The study utilized two main datasets:

  • Phantom Dataset: A TIMES phantom with nine vials containing different concentrations of agarose gel and nickel chloride was used to simulate various tissue properties. This phantom served as a benchmark for quantitative accuracy and repeatability.
  • In Vivo Datasets:
    • Healthy Subjects: Data from seven healthy volunteers (6 males, mean age 35.4 ± 3.6 years) were acquired to assess in vivo performance.
  • Patient Cohort: Data from three patients (3 males, mean age 63.0 ± 8.3 years) with suspected or known cardiovascular disease were collected for clinical feasibility assessment.

No new benchmark datasets were introduced; rather, existing quantitative MRI techniques (T1, T2, ECV, LGE) were used as comparators.

Key Insights and Ideas

  • Lower SAR, Higher Field Strength Viability: The TRAFF2 method offers a promising solution for myocardial tissue characterization at 3T by significantly reducing SAR, a major hurdle for T1ρ mapping at higher field strengths.
  • Sensitivity to Myocardial Injury: Preliminary clinical data suggest that TRAFF2 mapping can detect alterations in myocardial tissue consistent with infarction, potentially offering a non-contrast alternative to LGE.
  • Distinct Contrast Mechanism: TRAFF2 relaxation appears to probe different tissue properties than native T1 and T2, showing no significant correlation with T1 and a high correlation with T2 in phantom studies, suggesting it captures distinct molecular interactions.
  • Artifacts Remain a Challenge: While generally acceptable, TRAFF2 maps are susceptible to off-resonance artifacts, particularly around the coronary sinus, which impacts image quality and warrants further investigation for mitigation.

Key Results

  • Phantom: TRAFF2 showed excellent agreement with a reference method (R² = 1.00) and good sensitivity to phantom composition. Precision was excellent (average SD of 4.9 ± 2.3 ms), and intra-scanner repeatability was high (average CV of 1.4 ± 0.7%). TRAFF2 showed no correlation with T1 (R² = 0.09) but a high correlation with T2 (R² = 0.99).
  • Healthy Subjects: TRAFF2 mapping was successfully performed in a single breath-hold (≈17s) with acceptable image quality (IQ = 3.0 ± 1.0). Good precision (7.6 ± 1.4%), reproducibility (1.0 ± 0.6%), and low inter-subject variability (10.0 ± 1.8%) were achieved. Average TRAFF2 time was 79.1 ± 7.3 ms.
  • Patients: In two out of three patients with LGE-positive findings, TRAFF2 maps showed visual alignment with the scar tissue identified by LGE, suggesting potential sensitivity to myocardial infarction. Infarcted areas generally showed higher TRAFF2 values than remote areas.

Limitations and Future Work

  • Artifacts: Residual off-resonance artifacts, particularly near the coronary sinus, were observed and impacted image quality, necessitating further investigation into mitigation strategies.
  • Phantom vs. In Vivo: The phantom does not fully replicate tissue properties, which may explain the high correlation between TRAFF2 and T2 observed in phantom, which might not hold true in vivo.
  • Limited Patient Cohort: The clinical feasibility was assessed in a small proof-of-principle cohort (n=3), and the study included different imaging views for patients, limiting direct comparisons.
  • Lack of Histological Validation: Future research should include histological validation after biopsy to confirm findings.
  • Comparison with Other Methods: A direct comparison with other rotating frame relaxation methods like girRAFF and T1ρ was not performed.

Future work should focus on developing more inhomogeneity-resilient RAFF pulses (e.g., higher-rank RAFFn or girRAFF), exploring advanced shimming techniques, and evaluating TRAFF2 in larger patient cohorts to establish clinical sensitivity and determine cut-off values for distinguishing between remote and infarcted myocardium. Inter-scan variability assessment over multiple days is also recommended.

Summary updated

Abstract

Myocardial <mml:math xmlns:mml="http://www.w3.org/1998/Math/MathML"> <mml:msub><mml:mrow><mml:mtext>T</mml:mtext></mml:mrow> <mml:mrow><mml:mi>RAFF</mml:mi> <mml:mn>2</mml:mn></mml:mrow> </mml:msub> </mml:math> quantification at 3 T was successfully achieved in a single breath-hold with acceptable image quality, albeit with residual off-resonance artifacts. Nonetheless, preliminary clinical data indicate potential sensitivity of <mml:math xmlns:mml="http://www.w3.org/1998/Math/MathML"> <mml:msub><mml:mrow><mml:mtext>T</mml:mtext></mml:mrow> <mml:mrow><mml:mi>RAFF</mml:mi> <mml:mn>2</mml:mn></mml:mrow> </mml:msub> </mml:math> mapping to myocardial infarction detection without the need for contrast agents, but off-resonance artifacts mitigation warrants further investigation.
BibTeX
@inproceedings{tourais2024feasibility,
  author = {João Tourais and Maša Božić‐Iven and Yidong Zhao and Qian Tao and Iain Pierce and Christian Nitsche and George Thornton and Lothar R. Schad and Thomas A. Treibel and Sebastian Weingärtner and Mehmet Akçakaya},
  title = {{Feasibility of relaxation along a fictitious field in the 2nd rotating frame (TRAFF2) mapping in the human myocardium at 3 T}},
  booktitle = {Frontiers in Cardiovascular Medicine},
  year = {2024},
  doi = {10.3389/fcvm.2024.1373240},
}
Figure