| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
Letters to the Editor |
Department of Diagnostic and Interventional Imaging, University of Texas Medical School at Houston, 6431 Fannin Street, MSB 2.100, Houston, TX 77030
e-mail: Khader.M.Hasan{at}uth.tmc.edu
The article by Dr McGraw and colleagues (1) published in the July 2005 issue of Radiology builds on a previous preliminary report (2) in this journal by the same group. These articles are the first to document the utility of quantitative diffusion tensor magnetic resonance (MR) imaging metrics in monitoring stem cell therapy in vivo in a rare human disease (Krabbe disease). In both articles, only fractional anisotropy (FA) has been used and reported as a marker of the tissue microscopic changes that result upon the administration of stem cells. Although anisotropy indices, and FA in particular, are sensitive markers of the tissue microstructural organization with preferable signal-to-noise properties (see reference 3 and references therein), FA is not a specific marker of disease or myelination (4–6); alternatively, the transverse translational water molecular diffusivity would have probed the alteration of axonal morphology (7–10). The authors in these two articles did not report the mean diffusivity or the tensor regional eigenvalues to help assess the directional changes in white matter. Thus, the FA index alone as reported in these articles could not be used by the readers without additional quantitative information not presented in the articles in the assertion that stem cells have helped axonal remyelination, which would be associated with an increase in FA but would be directly related to a decrease in the transverse eigenvalue. An increase in tensor anisotropy may not be directly associated with white matter myelination (11–13). Assuming axial symmetry, the availability of FA and mean diffusivity would have been sufficient to compute the eigenvalues as described elsewhere (3,14,15). A discussion of the relevance and utility of the individual tensor diffusivities to tissue maturation and disease can also be found in an article by Mukherjee et al (8), in the recent works of Song et al (see reference 7 and references therein), and in an article by Kinoshita et al (9).
In conclusion, we strongly suggest that such novel quantitative articles provide the regional mean diffusivity values in addition to scale-independent anisotropy values such as FA. This information would enable the estimation of the transverse and longitudinal eigenvalues if the tensor eigenvalues are not reported directly. The availability of such information would provide more specific hints on the contributors to anisotropy and increase the diagnosis and pathologic specificity and utility of these simple and noninvasive diffusion tensor MR imaging metrics in future multicenter clinical trials (6–10,16).
 |
References |
|---|
 TOP References References |
|---|
Department of Radiology, Duke University Medical Center, Box 3808, Durham, NC 27710-3808
I thank Dr Hasan for his interesting comments, which underscore a few points that I thought important in the article. Dr Hasan's comments allow me a chance to expand on those points.
One question that investigators in the field of diffusion tensor imaging face is exactly what information various indices (eg, FA) obtained from diffusion tensor imaging provide. At present, these matters are not completely resolved but are under active study. We are glad to see that Dr Hasan is of the opinion that FA values "are sensitive markers of the tissue microstructural organization." Therefore, according to Dr Hasan's letter, he would presumably not object if we used the following statement to summarize our article: "Our findings suggest that diffusion tensor imaging appears to appropriately reflect improvements in the tissue microstructural organization associated with stem cell transplantation." Such a statement is already very close to what we affirmed in the article. This statement adequately summarizes one of the important points of the article—that is, we were able to show that diffusion tensor imaging appears to be a robust method for following treatment changes in Krabbe disease. We hope that this technique will prove equally useful for other white matter disorders.
Dr Hasan notes that FA values are not sufficient for derivation of data that reflect myelination. We are aware that a number of factors exist that can influence FA values. Some investigators believe that axonal integrity may also influence FA values. For that reason, at multiple points in the discussion section of our article, we stated that the increases in FA seen in patients undergoing early treatment of Krabbe disease may be due to "progressive myelination and increase in axonal integrity." It is also possible that Dr Hasan's comments are meant to address the fact that changes in FA can be influenced by changes in mean diffusivity values. We are aware of this possibility and have carefully examined both FA values and apparent diffusion coefficient values in a recent article (1) in Radiology in the analysis of tumors. In that study, we found that degree of increase in apparent diffusion coefficient values in various peritumoral regions generally tended to correlate with degree of FA decrease. This finding suggests that even if apparent diffusion coefficient values can affect FA values, changes in one value often parallel the other.
Although it is possible that FA values are not sufficient for derivation of data that reflect myelination, certainly our experience is that changes in FA values do indeed reflect changes in myelination. Krabbe disease is one in which the oligodendroglia responsible for myelin formation are damaged or nonfunctional. In our study, higher FA values were seen in patients who were treated early in life for this abnormality. For various humanitarian and medical reasons, we cannot absolutely prove that myelination occurred in individual patients in our study. Nonetheless, it appears likely that the FA values reflected progression of myelination on the basis of the following logic. Krabbe disease is a disease of disordered myelination due to a decrease in galactosylceramidase activity. In an animal model of Krabbe disease (the so-called twitcher mouse), bone marrow transplantation has been shown to increase galactosylceramidase activity and diminish neurologic symptoms and signs, which would be expected to improve myelination in this dysmyelinating disorder (2). We treated individuals with the same dysmyelinating disease by using stem cell transplantation and saw clinical improvement, decreases in areas of hyperintense signal abnormality on T2-weighted images, and increases in FA in infants who were treated early after birth. To insist that this improvement is not due to myelination because apparent diffusion coefficients were not measured is (perhaps) technically correct but appears to miss the major clinical and research points of the article.
On a similar note, we have shown that age-related increases in FA closely correlate with the expected progression in changes in signal intensity on T1-weighted and T2-weighted images during normal childhood (3), which multiple investigators have thought to reflect progression of myelination (4,5). These signal intensity changes during the 1st year of life are now generally accepted to represent progression of myelination. Could there be factors other than myelination that cause changes in signal intensity during normal childhood?—unquestionably. Does it seem likely that increases in FA values during normal childhood parallel progression in myelination?—the answer also appears to be "yes."
|
References |
|---|
|
TOPReferencesReferences |
|---|
This article has been cited by other articles:
|
|
 |
|
  C. Yu, F. Lin, K. Li, T. Jiang, W. Qin, H. Sun, and P. Chan Pathogenesis of Normal-appearing White Matter Damage in Neuromyelitis Optica: Diffusion-Tensor MR Imaging Radiology, December 1, 2007; 246(1): 222 - 228. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| RADIOLOGY | RADIOGRAPHICS | RSNA JOURNALS ONLINE |
