Skin tissue showed enhancement after GBCA administration. This phenomenon was confirmed by profile curve analysis. Subtraction image between pre-post contrast was calculated after image registration. This subtraction image showed distinct contrast effect of the skin tissue and may be helpful for the clinical diagnosis.
In contrast MRI examination of the brain, the image obtained by subtracting the image before contrast from the post-contrast image, were useful for the detection of lesions showing enhancing effect by gadolinium based contrast agent (GBCA), such as metastatic brain tumor. Viewing this subtraction MR image in the maximum intensity projection (MIP) method, signal enhancement was observed in the skin tissue. Skin tissue appears to show contrast with GBCA. The purpose of this study was, first, whether the signal enhancement of skin tissue after GBCA administration is observed, and verified using profile curve analysis method. Then, the subtraction MR image using the MIP method was evaluated to examine the clinical usefulness of a characteristic MRI imaging findings indicated by the skin tissues other than the brain.
This study was carried out as a
retrospective study. MRI used in this study was as follows. 3-T or
1.5-T MAGNETOM TRIO, SKYRA, PRISMA and AVANTO (Siemens Healthcare GmbH,
Germany). Dedicated head coil was used in each MR examination. The
subject patient using MRI, imaging of the head region was performed, including
brain tissue and the surrounding skin tissue. Three-dimensional T1-weighted
image (MPRAGE or VIBE) were taken before and after GBCA administration.
Sagittal image was obtained in the same FOV size and position. Spatial resolution
was 0.9mm isotropic. Before creating the subtraction image, image registration
processing of the before and after contrast image was performed. GBCA in
this retrospective study the following five types were used. Gd-DTPA,
Gd-DTPA-BMA, Gd-HP-DO3A, a Gd-BT-DO3A and Gd-DOTA. GBCA of the standard dose in
Japan was used. Per body weight 1kg, it was 0.1mmol or 0.05mmol. For displaying
subtraction image, AquariusNet software (TeraRecon, Inc., USA) was used. Displaying
the entire head in the MIP image, then, changing the slab thickness to 5mm from
10mm, the whole head in the three directions of view was used. Evaluation of
subtraction images were performed by a radiologist with 20 years of experience.
The MRI images were evaluated for the following. Blood vessels of the brain
tissue, in particular deep vein and choroid plexus, of the organization of the
contrast effect, neoplastic lesions such as metastatic brain tumor, dura mater
tissue of the skull inner surface and of the entire head skin. Especially for
evaluation of the contrast effect of the skin, it was confirmed by performing a
profile curve analysis whether the signal increase in the skin area was
observed. Further, using the MIP image of a thin slab thickness, investigated
clinical utility with the head skin contrast effect.
1. Abraham JL, Thakral C. Tissue distribution and kinetics of gadolinium and nephrogenic systemic fibrosis. European journal of radiology. 2008 May;66(2):200-7. PubMed PMID: 18374532. Epub 2008/04/01. Eng.
2. Aime S, Caravan P. Biodistribution of gadolinium-based contrast agents, including gadolinium deposition. Journal of magnetic resonance imaging : JMRI. 2009 Dec;30(6):1259-67. PubMed PMID: 19938038. Pubmed Central PMCID: 2822463.
3. Barral JK, Bangerter NK, Hu BS, Nishimura DG. In vivo high-resolution magnetic resonance skin imaging at 1.5 T and 3 T. Magnetic resonance in medicine. 2010 Mar;63(3):790-6. PubMed PMID: 20146351. Pubmed Central PMCID: 2832104.
4. Bellin MF, Van Der Molen AJ. Extracellular gadolinium-based contrast media: an overview. European journal of radiology. 2008 May;66(2):160-7. PubMed PMID: 18358659.
5. Bittoun J, Querleux B, Darrasse L. Advances in MR imaging of the skin. NMR in biomedicine. 2006 Nov;19(7):723-30. PubMed PMID: 17075954.