Myocardial T2* mapping free of distortion using susceptibility weighted spin-echo based imaging: a feasibility study

Introduction

Myocardial T2* mapping is of proven value for the assessment of myocardial iron content and tissue oxygenation [1, 2]. Conventionally, T2*-weighting is accomplished with gradient echo based or echo planar imaging techniques. However, the disadvantages of T1-related saturation effects, artifacts due to ventricular blood flow and image distortion must be addressed to pave the way for a broader clinical acceptance. Hence, spin-echo based acquisition strategies which generate T2* contrast free of distortion represent a valuable alternative.

Purpose

This study demonstrates the promise of navigator gated, susceptibility weighted, fast spin-echo imaging in conjunction with ventricular black blood preparation, for anatomically accurate T2* mapping of the heart.

Methods

The proposed cardiac triggered imaging technique is comprised of (i) a double inversion recovery module for ventricular blood suppression (ii) a navigator module for respiratory motion compensation and (iii) a fast spin echo imaging module. T2*-weighting was accomplished by implementing displaced UFLARE [3] using an extra evolution time tau between the initial excitation pulse and the first refocusing pulse (Figure 1). Volunteer studies were conducted using a 5-element cardiac coil array at 1.5 T (Philips Achieva, Best, Netherlands). Images were acquired with: scan matrix 192 × 192, FOV 250 mm, echo train length 20, TR = 2 R-R intervals, 3 slices with a thickness of 8 mm each. T2*-maps were generated from a series of images with tau ranging from 0 to 40 ms.

Figure 1

Basic scheme of cardiac and navigator gated, black blood prepared, susceptibility weighted UFLARE.

Results

The images of the test object in Figure 2 demonstrate the geometrical integrity of the UFLARE image. Conversely, the EPI images revealed distortions of up to 1 cm. Figure 3 depicts short axis views of the heart derived from a healthy subject using free breathing UFLARE in conjunction with double IR preparation without (left) and with T2*-weighting (middle) using an evolution time of tau = 20 ms. Image quality, signal-to-noise ratio and ventricular blood suppression are suitable for clinical applications. Even for strong T2*-weighting the images are free of distortions due to B0-inhomogeneities and free of physiological motion artifacts. Figure 3 depicts a T2*-map (right) derived from a UFLARE data set which is superimposed to an anatomical image. T2* distribution in the myocardium is rather uniform as it is to be expected for a healthy subject. The mean T2* value of the myocardium was found to be 29.9 +- 6.6 ms which matches previous reports [4].

Figure 2

Geometrical fidelity of T ₂ *-weighted images overlaid on a contour plot of the test object (red): UFLARE (yellow) and EPI (green).

Figure 3

Short axis views obtained with UFLARE Left) without susceptibility weighting and middle) with strong T₂ *-weighting). Right) corresponding T₂ *-map calculated from multi τ UFLARE data. The color represents T₂ * values ranging from T₂ * = 1 ms (black) to T₂ * ≥ 85 ms (dark red).

Conclusion

The feasibility and anatomic fidelity of T2*-weighted fast spin echo imaging have been demonstrated together with the image quality advantage over EPI. The proposed susceptibility spin-echo based approach promises to extend the capabilities of CVMR, including mapping and quantification of myocardial iron content, assessment of endothelial function, detection of stress induced angina pectoris, and differentiation of arteries and veins, which have all been elusive hitherto. It also holds the promise to potentially obviating the need for contrast agents for the assessment of myocardial perfusion while avoiding the drawbacks of commonly used EPI and gradient echo based approaches. In conclusion, we anticipate the extension of this work to higher field strengths to maximally exploit the SNR and contrast-to-noise (CNR) advantage and (iii) to evolve towards three-dimensional, distortion free, blood oxygen level dependent (BOLD) imaging of the heart in concert with parallel imaging.