| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
Letters to the Editor |
* Department of Radiology, Santa Casa de Misericórdia de São Paulo, Rua Cesario Mota Jr, 61, 01221-020, São Paulo, SP, Brazil
e-mail: Antonio.rocha{at}fleury.com.br
Department of Radiology and Radiological Sciences, Vanderbilt University Medical Center, Nashville, Tenn
We read the article "Diffusion-Tensor MR Imaging of Corticospinal Tract in Amyotrophic Lateral Sclerosis and Progressive Muscular Atrophy" (1) by Dr Cosottini and colleagues, in the October 2005 issue of Radiology, with great interest. We would like to use this opportunity to discuss important diagnostic and management issues of amyotrophic lateral sclerosis (ALS). The authors stated that several conventional magnetic resonance (MR) imaging techniques have been tested for the diagnosis of ALS and all of them have showed low sensitivity and specificity. The study by Dr Cosottini and colleagues (1) evaluated the use of a fluid-attenuated inversion recovery (FLAIR) sequence and showed that FLAIR imaging has limited value for the diagnosis of ALS, as shown previously (2). However, there are several published articles that show that magnetization transfer techniques do have a role in the diagnosis of ALS (2–4). We have shown that the T1-weighted spin-echo magnetization transfer contrast sequence is both sensitive and specific for the diagnosis of ALS, even in early stages of the disease (3).
Because of its rarity and inevitable lethality, the presumptive diagnosis of ALS is important. A technique that allows the early diagnosis of ALS is highly desirable. In this context, two types of examinations should be considered: (a) a simple and subjective test of the signal intensity of the corticospinal tract (CST) to rule in or rule out intracranial compromise of the CST and (b) a quantitative evaluation of the CST compromise for the longitudinal follow-up of patients. We believe that at earlier stages of the disease, when a definite clinical diagnosis of ALS is more difficult, it is preferable to use an examination that is simple to perform with images that are simple to interpret (3). On the other hand, quantitative examinations, such as the diffusion-tensor MR indexes and magnetization transfer ratio, may have a role in gauging tissue damage and disease progression, allowing the comparison of different case series and treatment follow-up of individual patients.
Results of our earlier published studies showed high sensitivity and specificity of white matter high signal intensity above the internal capsule on T1-weighted spin-echo magnetization transfer contrast images in the diagnosis of ALS, especially among patients with definite ALS. We are currently evaluating serial diffusion-tensor MR imaging (Signa Echospeed Excite II, GE Medical Systems, Milwaukee, Wis; 7500/90.3 [repetition time msec/echo time msec]; field of view, 26 x 26 cm; 5-mm-thick sections and no intersection gap; 128 x 128 matrix; two signals acquired; b = 1000 seconds/mm2; in 25 directions) in patients with definite ALS. Mean diffusivity and fractional anisotropy alone failed to help diagnose ALS in some of our preliminary cases when using those averages proposed by Dr Cosottini and colleagues. However, those same patients did have high signal intensity of the CST on T1-weighted spin-echo magnetization transfer contrast images, which is characteristic of ALS. We have noticed a tendency for concordance between abnormal signal intensity on T1-weighted spin-echo magnetization transfer contrast images and diffusion-tensor abnormalities, but this relationship will need to be confirmed with further studies.
We consider that the average of diffusion-tensor values, obtained only below the internal capsules, represents a limitation of the method proposed by Dr Cosottini and colleagues. However, comparing different series of patients with ALS is difficult because of the great individual variability of presentation and evolution. Therefore, the quantitative analysis of CST compromise may vary significantly from one patient to the other and even in the same patient over time. The comparison of mean fractional anisotropy and mean diffusivity values with those described by Dr Cosottini and colleagues may be impaired by many factors, such as individual differences in the extent and severity of CST compromise and even asymmetry of CST compromise. Moreover, the analysis of diffusion-tensor images may be affected by intra- and interobserver variability, since the region of interest in the CST within the internal capsule and brainstem is rather small and may be affected by partial volume averaging. Other technical differences may further impair comparison among different institutions and impair the reproducibility of different studies. Each institution would need to validate its results with a large number of affected individuals and control subjects in order to establish reliable parameters for ALS and normalcy, which would be a very difficult task given the low incidence of ALS.
Currently, the diagnosis of ALS is based on clinical and/or electromyographic criteria. MR imaging is used only to exclude diseases that could simulate ALS (5). However, new MR imaging techniques such as magnetization transfer and diffusion-tensor imaging may allow the diagnosis of ALS by showing the compromise of the intracranial segments of the CST. In this context, we believe that a quantitative technique that demands intensive postprocessing, such as is described by Dr Cosottini and colleagues, is not practical for many institutions and is less useful for individual patients. We believe that the use of MR imaging in ALS should be encouraged to expand the knowledge of the pathologic process of this disease in vivo. We believe that T1-weighted spin-echo magnetization transfer contrast imaging is useful and reliable for the early diagnosis of this disorder, while diffusion-tensor imaging is a reliable tool to estimate the structural damage that occurs with the progression of ALS, as shown by Dr Cosottini and colleagues.
 |
References |
|---|
 TOP References References |
|---|
,
Marco Giannelli, PhD,
and
Maria Chiara Michelassi, MD,
* Department of Neuroscience, University of Pisa, via Roma 67, 56100 Pisa, Italy
e-mail: mircocosottini{at}libero.it
Unit of Neuroradiology and Unit of Medical Physics, Azienda Ospedaliero-Universitaria Pisana, Pisa, Italy
Department of Oncology, Transplants, and Advanced Technologies in Medicine, University of Pisa, Pisa, Italy
We thank Dr da Rocha and colleagues for the interest in our work (1) and for their comments about the role of T1-weighted spin-echo magnetization transfer contrast imaging (2) and diffusion-tensor MR imaging in diagnosing ALS.
MR imaging can be used in ALS as a noninvasive tool to rule out other diagnostic possibilities. The use of conventional and advanced MR imaging techniques in diagnosing and studying ALS has been investigated in many works. Conventional MR imaging includes FLAIR and T1-, T2-, or intermediate-weighted imaging, whereas advanced and quantitative MR imaging includes spectroscopy, magnetization transfer ratio, and diffusion-tensor imaging. Conventional MR imaging has limited value in the diagnostic workup of ALS because of its low sensitivity and specificity in the detection of CST degeneration (3). MR spectroscopy is able to reveal a reduced concentration of the neuronal marker N-acetylaspartate in the motor cortex of patients with ALS, although quantitative methods need to be better standardized for clinical applications. Magnetization transfer ratio and diffusion-tensor imaging seem to be promising quantitative MR imaging techniques in the diagnosis and study of ALS. Magnetization transfer ratio is related to the magnetization transfer between bound protons associated with macromolecules and mobile protons associated to free water. Diffusion-tensor imaging involves the measurement of the diffusion tensor in tissue and uses mobile water molecules, during their random diffusion-driven displacements, to probe tissue structure at a scale that is well beyond image resolution (4).
The role of quantitative magnetization transfer ratio measurement in diagnosing CST impairment in patients with ALS was initially investigated by Kato et al (5). They found a significant decrease of magnetization transfer ratio (percentage) values in patients with definite ALS (mean ± standard deviation, 15.76% ± 1.48) in comparison with control subjects (19.83% ± 1.54). From the data reported by Kato et al, it can be inferred that an upper threshold magnetization transfer ratio value of 16.75 (2 standard deviations below the mean magnetization transfer ratio in control subjects) allows for a sensitivity of 77% and a specificity of 100%. Dr da Rocha and colleagues obtain similar results by using a qualitative analysis of magnetization transfer–weighted MR imaging (T1-weighted spin-echo magnetization transfer contrast) (2). The work of Dr da Rocha and colleagues suggests that, in patients with definite ALS, the decrease in magnetization transfer ratio value at the level of the CST can be related to a signal intensity increase (hyperintensity) detectable on magnetization transfer–weighted images. The T1-weighted spin-echo magnetization transfer contrast imaging technique represents a new qualitative, useful tool with high sensitivity (80%) and specificity (100%) in the diagnosis of ALS. This method can be easily implemented with many clinical MR imagers and is not time consuming since it does not need postprocessing steps.
Although the use of diffusion-tensor MR imaging as a marker to identify individuals with ALS at an early stage requires further investigation, there are several works that have analyzed diffusion-tensor imaging performances in the diagnosis of ALS (6–9). The diffusion-tensor MR imaging methods have a sensitivity and a specificity higher than do conventional MR techniques such as FLAIR or T1- and T2-weighted imaging. Graham et al (9), using high-angular-resolution diffusion-tensor imaging, report a sensitivity of 95% and a specificity of 71% in the detection of the upper motor neuron disease in patients with ALS. However, diffusion-tensor imaging is a technique that is sensitized to the diffusive properties of water molecules and offers the possibility of detecting some brain injuries earlier than they can be detected with other imaging methods (10). In this context, we point out that diffusion-tensor MR imaging has been shown to reveal early upper motor neuron involvement in patients with ALS before clinical symptoms of CST lesions become apparent (8).
The assessment of the performance of diffusion-tensor imaging in depicting CST involvement in ALS goes beyond the aim of our work (1). Diffusion-tensor imaging is a useful tool for studying the mechanisms underlying many pathologic processes. On the basis of these concepts, in our work we used diffusion-tensor imaging to investigate correlations between clinical variables and diffusion indexes to obtain further insight into the structural changes of the CST in patients with ALS and progressive muscular atrophy. The modifications in the diffusion-tensor eigenvalues observed in our study indicate that the underlying structural changes in ALS could be caused by nerve tract degeneration that resembles wallerian degeneration. To date, this result is unlikely to be obtained with MR techniques other than diffusion-tensor imaging.
Dr da Rocha and colleagues refer to a preliminary study comparing T1-weighted spin-echo magnetization transfer contrast with a diffusion-tensor imaging region-of-interest–averaged method in diagnosing CST compromise in patients with definite ALS. We believe that such study is interesting, and it could be useful to compare diffusion-tensor imaging and T1-weighted spin-echo magnetization transfer contrast imaging also in a longitudinal study that includes patients with an early, not definite form of ALS. We remark that diffusion-tensor imaging is a quantitative tool, and comparisons among different institutions have to be performed with caution (11,12) after a validation of the basic performances of the clinical MR imager (13). Moreover, a meaningful comparison of the diffusion-tensor imaging results obtained in different studies have to be made considering patients matched for disease duration and disease severity.
In their comments, Dr da Rocha and colleagues state that the average of diffusion-tensor values, obtained only below the internal capsules, represents a limitation. In this regard, since in regions with normally high anisotropy diffusion-tensor imaging parameters, changes are expected to be more pronounced and partial volume effects are less important, we decided to measure and average diffusion-tensor imaging parameters only below the internal capsules (a CST segment characterized by a high anisotropy due to high ordered white matter content). By using a region-of-interest–based analysis of diffusion-tensor imaging parameters, the partial volumes effects in CST regions with low anisotropy may reduce the sensitivity of diffusion-tensor imaging in depicting CST impairment in ALS. Really, in the study of Graham et al (9), where diffusion-tensor imaging parameters are measured at the level of the precentral gyrus, corona radiata, and internal capsule, the only significant changes of diffusion-tensor imaging parameters are revealed in the internal capsule. On the other hand the use of a voxel-based analysis of diffusion-tensor imaging parameters, probably for a lower partial volume effect, seems to be able to reveal diffusion-tensor imaging changes also in CST regions with low anisotropy, such as underneath motor and premotor cortex (8).
Magnetization transfer–weighted imaging and diffusion-tensor imaging represent complementary and promising MR imaging techniques in diagnosing and studying ALS. The T1-weighted spin-echo magnetization transfer contrast technique can be easily implemented on many MR imagers to corroborate the clinical diagnosis of ALS, while the quantitative measurement of diffusion-tensor imaging parameters can be used to monitor disease progression or therapy response and to study the underlying pathologic process.
|
References |
|---|
|
TOPReferencesReferences |
|---|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| RADIOLOGY | RADIOGRAPHICS | RSNA JOURNALS ONLINE |
