Dark Blood Late Gadolinium Enhanced Imaging of Myocardial Scar using First-Moment-Nulled Motion Sensitized Driven Equilibrium (m2MSDE)
Gregory J Wilson1, Niranjan Balu1, Jinnan Wang1,2, Chun Yuan1, and Jeffrey H Maki1

1University of Washington, Seattle, WA, United States, 2Bayer Healthcare, Whippany, NJ, United States

Synopsis

A novel black-blood pre-pulse is described that darkens intraventricular blood pool signal in late gadolinium enhanced (LGE) imaging of myocardial scar. The pre-pulse is m1-nulled motion-sensitized driven equilibrium (m2MSDE) with user-specified motion-sensitizing direction. The pre-pulse nulls blood signal while maintaining good myocardial image quality. Preliminary results are described.

Introduction

Cardiac late gadolinium-enhancement (LGE) MR is an important clinical tool for evaluating myocardial infarction and infiltrative cardiomyopathies. Traditionally, LGE-MR is performed 10-20 minutes after administration of gadolinium contrast using a 2D inversion recovery (IR) turbo gradient echo sequence with appropriate inversion time to null normal myocardium and highlight scar. Unfortunately, blood pool also appears bright on LGE, particularly when using phase sensitive IR (PSIR) techniques where TR is doubled. This can decrease scar conspicuity, particularly for subendocardial lesions bordering the blood pool. A novel black-blood pre-pulse was developed to provide black-blood LGE myocardial imaging that preserves myocardial image quality while improving depiction of subendocardial scar.

Method

First-moment-nulled motion-sensitized driven equilibrium (m2MSDE, Figure 1) was implemented as a black-blood prepulse (~20 ms duration) to the standard IR 3D or 2D gradient-echo LGE (non-PSIR) imaging sequence on a 3T whole body scanner (Ingenia, Philips, Best, the Netherlands). The pre-pulse utilized a motion-sensitized gradient scheme to null the first gradient moment, thus reducing sensitivity to bulk motion and improving myocardial image quality compared to standard MSDE. In addition, the direction of the m2MSDE gradients were specified in order to minimize the impact on myocardial image quality. For example, when imaging a short axis slice, the majority of myocardial motion is in-plane, so the m2MSDE gradients were applied through-plane. This leaves bulk motion unaffected, while intraventricular blood is nulled by intravoxel dephasing during the m2MSDE prepulse.1 To maximize pre-pulse effectiveness, the readout was performed with a low-high acquisition order and minimal (usually 2) startup echoes. The amplitude and duration of the m2MSDE gradients and the cardiac trigger delay were varied to balance the effects of blood suppression and myocardial image quality degradation. With IRB approval, m2MSDE LGE has been applied to 2 volunteers (without Gd) and 8 patients undergoing conventional LGE.

Results

Using an IR-TFE sequence without contrast agent, good blood suppression can be obtained with m2MSDE (Figure 2). The optimal combination of quiescent myocardial motion and high blood flow occurs at the beginning of diastole, so best results were obtained when trigger delay was set (user defined) to early diastole as determined from cine images. m2MSDE blood suppression is more difficult post-contrast, as the short T1blood leads to rapid recovery of dephased moving spins in the blood pool (Figure 3). For post-Gd suppression, a low-high k-space acquisition was used with shot length limited to approximately 100 ms. Blood is better suppressed in the high-flow base of the ventricle than in the apex where flow is limited – particularly in patients with reduced cardiac function. Figure 4 shows m2MSDE blood suppression in a patient with myocardial scar. In this case, blood was adequately suppressed in the basal portion of the left ventricle, but not suppressed near the apex. Fortuitously, the scar was well depicted as it was located in the basal portion of the LV.

Discussion

Other studies demonstrate MSDE can suppress enhanced flowing blood in the arteries.1 However, myocardial motion and highly variable blood flow patterns increase the challenge of applying the technique to cardiac imaging. Particularly in ill patients, myocardial motion is not completely eliminated by cardiac triggering to diastole, and degradation of myocardial image quality cannot be tolerated in LGE imaging. m2MSDE was developed to be less sensitive to bulk motion, and early evaluation demonstrates m2MSDE can suppress both non-enhanced and enhanced blood in the heart without compromising myocardial image quality. Achieving significant suppression of Gd-enhanced blood is challenging, and requires low-high k-space acquisition with minimal startup echoes and relatively short shot duration. Suppression is particularly difficult in the apex where blood flow is slower. Efforts to optimize blood suppression in the apex are ongoing.

Acknowledgements

NIH 1R21EB017514-01A1 and Philips Healthcare

References

1) Wang J, Yarnykh VL, Yuan C. J Magn Reson Imaging 31(5):1256–63 (2010).

Figures

Figure 1. m2MSDE is applied prior to the turbo gradient-echo readout (T1-TFE) in this respiratory-navigated IR-TFE. Motion-sensitizing gradients (GMSDE) can be applied in any direction, and have zero first-moment (m1) so constant velocity spins do not accumulate phase. First-moment-nulling and GMSDE aligned with the long-axis of the LV help preserve myocardial image quality.

Figure 2. Good blood suppression and myocardial image quality using m2MSDE in a normal volunteer without contrast agent. TI was set below the myocardium null-point to allow visualization of normal myocardium for testing purposes. m2MSDE parameters are indicated in the figure. Increasing GMSDE duration improved blood suppression while maintaining good myocardial image quality.

Figure 3. Application of m2MSDE in a normal subject with contrast agent. A) GMSDE amplitude = 0, B-D) GMSDE amplitude = 18 mT/m, with varying inversion times TI. Note good blood suppression with TI-dependent myocardial and background suppression, and in the absence of scar, very little signal remains at signal-nulling TI.

Figure 4. Initial application of m2MSDE in patient with myocardial scar (2D, 4-chamber view). In standard PSIR sequence (left), scar (arrow) and blood pool are relatively iso-intense. With m2MSDE IR-TFE (right), blood pool is suppressed, improving scar visualization. Blood pool signal was effectively suppressed near the left ventricle base, but not well suppressed near the apex.



Proc. Intl. Soc. Mag. Reson. Med. 24 (2016)
0138