Gadoxetic acid enhanced liver MRI has excellent utility for diagnosing hepatocellular carcinoma (HCC), distinguishing hypervascular pseudolesions from small HCCs, and facilitating the discrimination between focal nodular hyperplasia and hepatic adenoma^1,2.

Recently, a new sequence combined with DIXON-based fast suppression technique, called mDIXON-enhanced T1 High-Resolution Isotropic Volume Examination (mDIXON-eTHRIVE) has been reported that this sequence showed improved homogeneity of fat suppression and higher overall image quality compared with conventional spectral inversion recovery-based eTHRIVE in contrast-enhanced liver MRI³. However, it often suffers from ghosting flow artifacts from aorta due to increased signals by contrast enhancement. Furthermore, flow-induced phase shift may potentially be cause misallocation artifacts⁴.

To solve this problem, we attempt to use motion-sensitized driven-equilibrium (MSDE)^5,6, which has previously been used for blood-suppression technique in the carotid⁵, intracranial⁷ and peripheral nerves⁸, for reducing through-plane flow artifacts particularly in dynamic contrast-enhanced studies. The purpose of this study was to compare two types of MSDE pulses that have been proposed so far and to evaluate whether MSDE can reduce the artifacts in contrast enhanced studies.

Fig.1 shows the scheme of two types of MSDE pulses. The original MSDE preparation consists of T2 preparation (90°/180°/-90°) pulses with motion- sensitizing gradients sandwiched between the RF pulses⁵. iMSDE⁶ has been developed to address the sensitivity of MSDE to eddy currents and the inhomogeneity of MSDE image quality. The iMSDE preparation consists of a 90° excitation pulse, two 180° Malcom-Levitt (MLEV) refocusing pulses, and a 190° flip-back pulse with motion-sensitizing gradients sandwiched between the RF pulses. We compared both in this study.

A total of 12 patients who underwent contrast-enhancement were examined on 3.0T whole-body clinical systems (Ingenia, Philips Healthcare). The study was approved by the local IRB, and written informed consent was obtained from all subjects.

All patients underwent 3 sequences (conventional mDIXON-FFE, mDIXON-TFE with MSDE, and mDIXON-TFE with iMSDE) with randomized acquisition order in between dynamic study and hepatobiliary phase (4~10min after contrast enhancement) to minimize the timing bias in estimating the signal intensity of the vessel signal.

To evaluate the effect of the MSDE or iMSDE in suppression of ghost artifacts quantitatively, we measured the contrast-ratio (CR) between the aorta and the subcutaneous fat. Theoretically the fat should be less affected by MSDE, which is based on T2prep, than muscle because this is based on Dixon sequence). The CR was assessed by using one-way repeated measures ANOVA and post-hoc Tukey test. Subsequently, we visually assessed and compared the presence of ghosting artifacts and overall image quality including SNR and contrast in all acquired images of three sequences.

Table 1 shows the results of the comparison of CR. Both mDIXON-TFE MSDE and iMSDE showed significantly higher CR compared to conventional mDIXON-FFE. mDIXON-TFEMSDE indicated slight higher CR compared to mDIXON-TFE iMSDE, but there was no significant difference.

In visual evaluation, both mDIXON-TFE MSDE and iMSDE could reduce the presence of ghost artifacts (Fig.2). The incidence of artifacts in both methods was significant lower than that of conventional mDIXON-FFE, but there was no significance between mDIXON-TFE MSDE and iMSDE. On the other hand, signal-to-noise ratio and image contrast of mDIXON-TFE MSDE images was visually higher than those of iMSDE images because preparation time of MSDE (10ms) significantly shorter than that of iMSDE (40ms). However, severe signal inhomogeneities on the liver could be observed in the images of MSDE in many cases (8/12 cases) compared to iMSDE (5/12 cases) probably due to higher sensitivity to field inhomogeneities of MSDE (Fig.3). Hence, iMSDE would be better for clinical studies in current sequence; however, further technical improvements for MSDE would be needed.

In this study, we showed the effect of MSDE in reducing through-plane flow artifacts particularly in dynamic contrast-enhanced studies. Both MSDE schemes could significantly decrease flow signals and could therefore reduce flow artifacts sufficiently. In current sequence, iMSDE would be better for clinical studies because of its less sensitivity to field inhomogeneities.