The aim of this study was to determine a moderate flip angle (FA) for dynamic liver MR imaging (MRI) with the three-dimensional volumetric interpolated breath-hold examination (3D-VIBE) technique. Images of phantoms with various T(1) values (44-560 msec) were acquired with the 3D-VIBE sequence (TR/TE=5.2/2.6 msec) using different FA (5-50 degree). We estimated signal-to-noise ratio (SNR) and contrast-to-noise ratio (CNR), considered to indicate tumor-to-liver contrast, as a function of FA. In phantoms, in which T(1) values (44-191 msec) were assumed to be shortened by the effect of Gd-DTPA in hepatocellular carcinoma (HCC), the highest SNR in each phantom was observed at FA ranging from 15 to 30 degrees. SNRs in other phantoms, in which T(1) values (298-560 msec) were assumed to be normal liver-tissue pre- or post-enhancement, were high with FA of 10-12 degrees, and were remarkably decreased with FA of more than 30 degrees. CNR increased as FA increased in every phantom, especially in the phantom with shortened T(1) values (44-191 msec), suggesting that enlarging FA improved the tumor-to-liver contrast. Taking both results of SNR and CNR into account, we concluded that a moderate FA was approximately 25 degrees. The advantage with an FA of 25 degrees was confirmed in a clinical study of a patient with hypervascular HCC, in which we could observe coronal enhancement surrounding the lesion in the late phase of the double arterial phase by dynamic MRI using the 3D-VIBE technique.

No abstract

In our study of three-dimensional contrast-enhanced MR angiography, we performed a computer simulation to quantitatively investigate vessel visibility according to scan timing. To construct the simulated MR images, we varied the position (scan timing) and range (enhancement-duration) of k-space data assumed to be acquired during contrast enhancement. In the present study, either the sequential or centric phase-encoding order in k(y) and k(z) on k-space was assumed to be used. When scan timing was shifted from the optimal timing, the visibility of thick vessels decreased, and the signal intensity in thin vessels was higher than that in thick vessels. We found that the appropriate setting of scan timing was an important factor in the visibility of thick vessels. Meanwhile, we also noted that extending the enhancement-duration (or shortening the scan time) could increase the visibility of thin vessels. Our results and the simple technique used for simulation are considered to be useful for the study of three-dimensional contrast-enhanced MR angiography.

The aim of the present study was to evaluate the increase in signal intensity caused by applying the fast recovery (FR) technique to fast spin echo (FSE) images, that is, the fast recovery fast spin echo (FR-FSE) method. All images of phantoms, whose T(2) values were different, were acquired with a Signa 1.5 Tesla system (GE Medical Systems) using the three-dimensional (3D) FSE and 3D FR-FSE sequences. We assessed the increased signal intensity as follows: (signal intensity on the FR-FSE image - FSE image) / FSE image (%). Our results showed that the increased signal intensity became high when 1) T(2) of the phantom was prolonged, 2) TR was shortened, and 3) echo train length (ETL) was decreased. By utilizing the results of this study, the increased signal caused by the FR technique could be estimated quantitatively when the TR, ETL, and T(2) of investigated substances were determined. For example, when TR, ETL, and T(2) were 1500 msec, 16-64, and 1500 msec, respectively, the increase in signal intensity was estimated to be approximately 70%. In addition, when T(2) was less than approximately 250 msec, signal intensity was not significantly increased by the FR pulses, that is, the FR-FSE image was the same as the FSE image. Accordingly, the FR-FSE method was confirmed to enhance the signal in substances with longer T(2), while maintaining the same contrast of the image as that obtained by the conventional FSE method. Our results are useful for evaluating the increased signal intensity caused by employing the FR technique.

In the present study, we quantitatively investigated the relationship between the signal intensity in a vessel and the duration of contrast enhancement as well as scan timing in 3D contrast-enhanced MR angiography using an elliptical centric phase-encoding technique. A tube phantom filled with Gd-DTPA, acting as a vessel, was taken out from the field of view during data acquisition, by using the "pause" function of our MR scanner (GE Signa, 1.5 Tesla), thereby simulating the presence and absence of a vessel. The shortening of the duration of enhancement corresponds to the delay of scan timing from the optimal point in the phase-encoding of the centric-ordering system. The signal intensity in a vessel (1-5 mm in diameter) decreased as the duration of enhancement became shorter and the diameter of the vessel decreased. When the number of partitions was 16 or 32 in a 128-mm-thick slab, the signal intensity obtained by the elliptical centric phase-encoding technique was almost the same as that obtained by the conventional centric phase-encoding technique. However, when the number of partitions was increased (64-124), and if the duration of enhancement was short, the signal intensity obtained by the elliptical centric phase-encoding technique was higher than that obtained by the conventional centric phase-encoding technique. In conclusion, in terms of the duration of enhancement and the delay of scan timing, the elliptical centric phase-encoding technique is superior to the conventional centric phase-encoding technique when the number of partitions in a slab for 3D MR angiography is increased.