Table 1.
The table represents of the most important sequence parameters of the standard liver imaging protocol as well as of the (phase-sensitive) inversion recovery sequence. To directly compare VIBE and IR sequences, parameters such as FOV, image matrix and slice thickness were identical for both sequences.
Fig 1.
Overall image quality and subjective delineation and visual contrast.
(A) Overall image quality as rated by readers and mean. Whiskers indicate 95% confidential interval. (B) Subjective delineation and visual contrast. The score describing subjective delineation and visual contrast was higher on MAG and PSIR sequences compared to VIBE in all size groups; overall, there was a statistically significant difference for VIBE vs. MAG vs. PSIR (p < .001).
Fig 2.
Axial images in a 74-year-old man with gall bladder carcinoma acquired with (A) T1 VIBE (repetition time msec/echo time msec, 4.74/2.38), and (B,C) T1-weighted inversion recovery (912/1.13) by using (B) magnitude and (C) phase-sensitive reconstruction 20 minutes after administration of gadoxetate disodium.
Note the distinct conspicuity of the lesion in segment VII (arrowhead) and the clear delineation of the small lesions (white arrows). Five lesions which are distinctly delineated on inversion-recovery images (B1, C1) are hard to detect on VIBE (A1), which illustrates the high relative liver-to-lesion contrast.
Fig 3.
Images of a 61-year-old man with hepatic metastases of malignant melanoma.
Constant appearance of a large haemorrhagic lesion (white arrowheads) that had been treated by local ablation beforehand. Nevertheless, a new metastatic lesion (white arrow) was seen in segment VIII: compared to the VIBE sequence (d), the lesion is more sharply demarcated on PSIR images with magnitude reconstruction (e), as well as with phase-sensitive reconstruction (f).
Fig 4.
(A) Mean LLC as measured on VIBE and on magnitude-reconstructed IR (MAG) sequences. The difference was statistically significant (p <0.05). (B) Size measurements revealed a close correlation between the sizes of metastases between both sequence types with slightly larger diameters on PSIR images in comparison to VIBE images. Typically, MAG values were slightly higher than VIBE values. The average difference was 1.30 mm.
Fig 5.
Overall number of metastatic lesions detected in the patient cohort.
PSIR images with magnitude reconstruction (black) visualised more lesions with a size between 0.5 and 1 cm and especially below 0.5 cm.
Fig 6.
Metastatic tumour in an 80-year-old man with central cholangiocellular carcinoma with metastatic spread.
Transverse T1-weighted VIBE (A), and T2 TSE-weighted (B) images. Dynamic VIBE imaging started 15 seconds after intravenous administration of gadoxetate disodium (C). (D) shows VIBE images 20 minutes after contrast agent administration, (E) inversion-recovery with magnitude reconstruction and with phase-sensitive reconstruction (F). The subcapsular lesion is more clearly depicted (white arrow) than on conventional VIBE images. Additionally, several lesions within the liver parenchyma affected by long-standing cholestatic disease can be clearly delineated on inversion recovery images, whereas reduced gadoxetate disodium uptake diminished delineation on conventional VIBE images (arrowheads).