[Mristudio-users] question - changes of FA and eigenvalues after Gd contrast

[susumu mori]

Hi Amir, nice observation.

For NMR/MRI of living organs, we are dealing with a system with a huge
number of compartments. Waters inside cytoplasm, nucleus, golgi, mito,
extracelluar space, bound to macromolecules, inside an axonal process,
trapped between neurofilaments/microtubles, inside myeline, etc. These
microscopic environment varies within a scale of micrometer scale or less.
Then, there are larger scale structure units such as neurons, astrocytes,
then even larger such as WM and GM, from small (tens of microns) to large
(millimeter). These structural units with different scales coexist within
one pixel.

In this complicated multi-compartment systems, water molecules move around
about 10 micron, in average, experiencing different environments, some place
could be isotropic and some could be anisotropic. If we can assume that
during the 40-60ms of observation time (typical separation between two
gradient pulses, in which water moved 10micron), all water molecules
experience all compartments and effectively averaged out the existence of
different diffusion environments, we can assume that the system is
homogeneous (fast-exchange regime). Usually, we can's assume it; many water
molecules are trapped in one compartment or visiting only several
compartments. Then our pixels consist of multiple compartments with
different diffusion environment.

Now, it is not only diffusion properties that are different among these
compartment. Their relaxation properties are also very different. Suppose
one simple and extreme case in which we have only two compartments and there
is no exchange between them (like intra and extra-cellular spaces with very
slow membrane exchange). Also assume that intra cellular component is
dominant (90%) with high anisotropy and short T2 and extra cellular
component is minority (10%) with low anisotropy and long T2 (I just made up
all these numbers). If we use a short echo sequence, the signal from the
intracelluar compartment dominate the signal and the system looks very
anisotropy. If we use a long echo sequence, the signal from the
extracellular space becomes dominant and the same pixel now looks less
anisotropic.

So, as long as we are dealing with multiple compartment systems with slow
exchange regime, changing relaxation time (like altering T1 and T2) and
sequence (like long echo or rapid repetition), you can effectively change
the final contribution of of each compartment to the signal. Therefore, it
is likely that T1, T2, AND "measured" diffusion properties change by adding
Gd, even if Gd doesn't directly change water diffusion.



On Sat, Jan 29, 2011 at 12:43 PM, Amir Zolal <amirzolal at gmail.com> wrote:

> Hi,
>
> I have tried comparing DTI images before and after Gd contrast
> administration, and I have found that the FA was increased after the
> contrast agent (more in the ROIs where there was contrast
> enhancement), I have also (among other things) seen a significant
> decrease of all eigenvalues, however the third eigenvalue decreased
> the most (the difference between the change of lambda1 and lambda3 was
> also significant), causing the paradoxical increase in FA. This was
> not observable in the normal white matter. Because I don't believe,
> that the contrast agent would in any way influence water diffusion, my
> theory is, that the addition of a paramagnetic contrast agent  could
> cause more local noise and the small attenuation in the (lambda3)
> direction would be even less discernible, in contrast to the strong
> attenuation in the (lambda1) direction. My question is, to your
> knowledge, would there be anything else in the processing of the DTI
> datasets that could cause the above described effect?
>
> Thanks in advance for your ideas,
>
> Amir
> _______________________________________________
> mristudio-users mailing list
> mristudio-users at mristudio.org
> http://lists.mristudio.org/mailman/listinfo/
> Unsubscribe, send a blank email to:
> mristudio-users-unsubscribe at mristudio.org
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://lists.mristudio.org/pipermail/mristudio-users/attachments/20110129/2f244142/attachment.html