| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
Pediatric Imaging |
1 From the F. M. Kirby Research Ctr for Functional Brain Imaging, Johns Hopkins University, Baltimore, Md (F.S.E., R.I., P.B.B., E.R.M., S.M., P.C.M.V.Z.); the Dept of Neurogenetics, Kennedy Krieger Institute, Baltimore, Md (F.S.E., H.W.M., G.V.R.); and the Dept of Radiology (R.I., P.B.B., E.R.M., S.M., P.C.M.V.Z.) and Oncology Center, Division of Biostatistics (E.S.G.), Johns Hopkins Univ School of Medicine, Baltimore, Md. Received Jun 13, 2001; revision requested Aug 6; final revision received March 12, 2002; accepted Apr 15. Supported in part by grants RR 00052 and HD 10981 from the U.S. Public Health Service. Address correspondence to E.R.M., Dept of Radiology, Hospital of the Univ of Pennsylvania, 3400 Spruce St, Ground Floor Founders Building, Philadelphia, PA 19104 (e-mail: emelhem@rad.upenn.edu).
 |
ABSTRACT |
|---|
 TOP ABSTRACT INTRODUCTIONMATERIALS AND METHODS RESULTSDISCUSSIONREFERENCES |
|---|
MATERIALS AND METHODS: Multisection proton MR spectroscopy and DT imaging were performed in 11 patients with X-ALD and in 11 healthy control subjects. Quantitative measures of N-acetylaspartate (NAA), choline, and creatine values and of isotropic apparent diffusion coefficient (IADC) and fractional anisotropy (FA) were obtained from coregistered regions of interest. DT imaging and metabolic parameters were compared by using regression analysis. In addition, differences in DT imaging and metabolite measurements between normal- and abnormal-appearing white matter on conventional MR images were evaluated by using a nonparametric (Mann-Whitney) test.
RESULTS: A strong logarithmic relationship between NAA value and FA (r = 0.64, P < .001) and an inverse logarithmic relationship were found between NAA value and IADC (r = -0.69, P < .001). Creatine and choline values correlated poorly with IADC and FA. In the normal-appearing white matter of asymptomatic patients, the NAA value was 17% lower than that in the healthy control subjects (P = .016), whereas no significant difference in DT imaging measures was seen in these regions.
CONCLUSION: In patients with X-ALD, MR spectroscopic imaging can depict abnormalities in white matter that have a normal appearance on both conventional MR and DT images; this finding suggests that it may be the most sensitive technique for detecting early abnormalities of demyelination or axonal loss in patients with X-ALD.
© RSNA, 2002
Index terms: Adrenal gland, abnormalities, 86.549 • Adrenal gland, MR, 86.121413, 86.12143, 86.12144, 86.12145 • Magnetic resonance (MR), diffusion tensor, 86.121419 • Metabolism
|
INTRODUCTION |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODS RESULTSDISCUSSIONREFERENCES |
|---|
N-acetylaspartate (NAA) is considered to be a neuronal marker (4–7) and is predominantly present in the axons within white matter. The signal from choline and creatine consists predominantly of glycerophosphocholine, phosphocholine, and free choline (8), compounds involved in membrane synthesis and degradation. Creatine, a composite signal consisting predominantly of creatine and phosphocreatine, is considered to be a measure of cellular density and is especially high in glial cells (9). The determinants of water diffusion properties within white matter are more controversial (10–13) and have been suggested to be influenced by the presence of myelin and axonal structure. The purpose of our study was to compare conventional MR, proton MR spectroscopic, and DT MR imaging findings in patients with X-ALD. None of the presented data have previously been published.
|
MATERIALS AND METHODS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODS RESULTSDISCUSSIONREFERENCES |
|---|
Conventional MR, Proton MR Spectroscopic, and DT MR Imaging
Conventional MR, proton MR spectroscopic, and DT MR imaging were performed with a 1.5-T clinical system (ACS-NT; Philips Medical Systems, Best, the Netherlands) with a maximum gradient capability of 23 mT · m-1 and a slew rate of 103 mt · m-1 · msec-1, with use of a quadrature head coil operating in transmit-receive mode. Conventional MR imaging consisted of acquisition of sagittal T1- (repetition time [TR] msec/echo time [TE] msec, 587/20) and transverse T2- (3,000/30–100) weighted and fluid-attenuated inversion recovery (6,000/58–160/2,000 [TR msec/TE msec/inversion time msec]), or FLAIR, brain images. According to our standard protocol, three patients who showed AWM at conventional MR imaging (patients 7, 8, and 10) intravenously received 0.01 mmol per kilogram of body weight of gadopentetate dimeglumine (Magnevist; Berlex Laboratories, Montville, NJ) after completion of proton MR spectroscopic and DT imaging. None of the patients required sedation.
For proton MR spectroscopic imaging, four oblique-transverse sections (nominal thickness, 15 mm; intersectional gap, 3 mm) were selected parallel to the anterior commissure–posterior commissure line. Outer volume saturation pulses were used to suppress lipid and water signals originating from the skull and scalp, and a chemical shift-selective saturation pulse was used for water suppression. The four interleaved sections were recorded (2,300/272; field of view, 24 cm; matrix size, 32 x 32; one signal acquired), resulting in a total of 30 minutes for data acquisition. The nominal voxel size was 0.8 cm3. Full technical details are given elsewhere (15).
For DT imaging, a total of 20 5-mm-thick sections with a 1-mm gap were obtained by using a segmental spin-echo echo-planar readout with cardiac gating (18 echoes per TR: one navigator echo and 17 phase-encoded echoes). Oblique-transverse images were prescribed parallel to the anterior commissure–posterior commissure line in the same oblique plane as proton MR spectroscopic images, with locations chosen carefully so that one MR spectroscopic section exactly matched three DT sections. Acquisition parameters were 4,286–5,600 (TR msec, depending on the subject’s heart rate)/94 (effective TE msec), field of view, 24 cm; matrix, 128 x 128, with an interpolated readout to 256 steps. At each level, three or four sets of DT images applied in six directions (x, y, z, x + y, x + z, and y + z); a b value of 599 sec/mm2 (gradient duration, 26.8 msec; separation time, 36 msec; gradient strength, 20.9 mt · m-1), and images were obtained without diffusion sensitization. The approximate acquisition time for a set of DT images was 4–5 minutes, for a total DT imaging time of 15–20 minutes.
Data Processing
For proton MR spectroscopic analysis, data were transferred to a workstation (Enterprise 5500; Sun Microsystems, Mountain View, Calif), at which data processing of each individual section was performed. Cosine filtering was applied in the k-space domains, giving an effective voxel size approximately double the nominal size. Peak areas were estimated in the frequency domain, with assumption of Gaussian line shapes made by using a simplex routine (16). Concentrations of total NAA, total creatine, and total choline were estimated by using the phantom replacement technique (16).
For DT imaging analysis, numeric calculations and image display were performed by using in-house routines written in Interactive Data Language (Research Systems, Boulder, Colo). Each set of DT images was reviewed by two readers (F.S.E., R.I.), motion-degraded images were rejected by means of consensus to improve signal-noise ratio, and the remaining motion-free sets of images were averaged together prior to tensor calculation. The six independent variables (Dxx, Dyy, Dzz, Dxy, Dyz, Dxz) in the DT were calculated from the DT images (17), where D represents diffusion coefficient. From the DT data, pixel-by-pixel brain maps were generated for isotropic apparent diffusion coefficients (IADCs) [(Dxx + Dyy + Dzz)/3] and fractional anisotropy (FA) (18). Mean IADC and FA values were calculated for each nominal 0.8-mL MR spectroscopic voxel (8 x 8 x 3 = 192 DT imaging pixels).
Classification and Data Analysis
Conventional (T2-weighted and FLAIR) MR images were used to assess cerebral white matter involvement in the patients with X-ALD. Two readers worked together and classified voxels as normal-appearing white matter (NAWM) or AWM by means of consensus on the basis of conventional MR images. Voxels with cerebrospinal fluid contamination were omitted. To account for regional variations related to different histopathologic zones in the patients with X-ALD, as well as for differences in the phenotypes of the patients with X-ALD, we took three approaches:
First, to evaluate regional variation in the white matter of the healthy control subjects and in NAWM and AWM in the patients with X-ALD, we selected columns of white matter voxels in the centrum semiovale, in which we performed a voxel-by-voxel analysis of all data (analysis 1). Second, we directly compared all NAA, creatine, and choline values with IADC and FA values within white matter, irrespective of the location of the voxels (analysis 2). Third, we investigated differences between mean NAA, creatine, and choline values and mean IADC and FA values for different patient subgroups (analysis 3).
For analysis 1, for each NAA, creatine, and choline value and for IADC and FA values, at each of the nine voxels sampled, the mean and standard error of the mean were estimated in the healthy subjects and in patients who were asymptomatic (had a mutant ALD gene without neurologic or endocrinologic deficit) or had Addison-only X-ALD (adrenocortical insufficiency without neurologic involvement). These were plotted in box plots and indexed by voxel location to assess whether there were any trends over voxels that were the same or differed in the two subject populations (Fig 1). A similar approach was taken to look at the results in one subject (Fig 2). Metabolite concentrations for right and left hemispheres were plotted on the same graphs to assess relative concentration shifts.
|
|
For analysis 3, we investigated differences in NAA, creatine, and choline values and in IADC and FA values in patient subgroups, described subsequently, as compared with those in the healthy control subjects. These differences were tested by using a nonparametric (Mann-Whitney) test. The three patient subgroups considered were NAWM in the patients who were asymptomatic or had Addison-only X-ALD, and AWM and NAWM in the patients with CCALD, ACALD, or X-ALD. All statistical analyses were performed by using Stata version 6.0 statistical software (Stata Corporation, College Station, Tex), and data were calculated by using Excel software (Microsoft, Seattle, Wash).
|
RESULTS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODS RESULTSDISCUSSIONREFERENCES |
|---|
|
|
The findings in the patients with X-ALD and AWM at conventional MR imaging are exemplified on the image in Figure 2 (patient 7), which was obtained in a 7-year-old patient with CCALD who had gait abnormalities and right hemiparesis. Conventional T2-weighted imaging showed an extensive bilateral white matter lesion in the parieto-occipital region that had greater involvement on the right than on the left. NAA, creatine, choline, IADC, and FA values in the right and left hemispheres are plotted in Figure 2. In both hemispheres, NAA decreased continuously toward the center of the parieto-occipital lesion. Creatine and choline increased in NAWM toward the edge of the lesion and decreased in AWM toward the center of the lesion. In the corresponding voxels, IADC values increased and FA values decreased continuously toward the center of the parieto-occipital lesion.
Direct Comparison of NAA, Creatine, and Choline Values with IADC and FA Values
In the healthy control subjects, the correlation coefficients of the NAA and IADC values and the FA values were 0.25 (P = .55) and -0.19 (P = .25), respectively. The correlation coefficients of the creatine and IADC values and the FA values were 0.03 (P = .53) and -0.24 (P = .33), respectively. The correlation coefficients of the choline and IADC values and the FA values were 0.30 (P = .01) and -0.16 (P = .78), respectively.
The relationship of NAA values with IADC and FA values in the patients with X-ALD is illustrated in Figures 3 and 4. Correlation coefficients for the relationships of NAA values with log(IADC) and log(FA) values in the white matter voxels were -0.69 and 0.64, respectively. After adjusting for the correlation between multiple observations in the same individual, the P value describing the association between NAA versus log(IADC) and NAA versus log(FA) was less than .001.
|
|
|
|
|
|
2. In NAWM, correlation coefficients for choline values versus IADC and FA values were 0.27 (P = .14) and -0.28 (P = .28), respectively (Figs 5, 6). Correlation coefficients for creatine values versus IADC and FA values were -0.20 (P = .57) and 0.08 (P = .66), respectively (Figs 7, 8).
Mean NAA, Creatine, and Choline Values versus IADC and FA Values in the Different Subgroups
Table 1 shows the mean metabolite concentration and diffusion property values and SDs in the healthy control subjects. Table 3 shows the differences in the patients with X-ALD as percentages of the mean values in the healthy control subjects:
|
2. In NAWM in the patients with CCALD or ACALD, NAA and creatine values were significantly lower than those in the healthy control subjects (P = .037 for both). There was also a small increase in choline value, but this was not significant (P = .296), at least not in this limited population.
3. In the patients with AWM, NAA and FA values were lower and IADC values higher than those in the healthy control subjects (P = .010 for all three variables). Choline showed a small increase, but again, this was not statistically significant (P = .392).
|
DISCUSSION |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODS RESULTSDISCUSSIONREFERENCES |
|---|
Spectroscopic changes in NAWM that precede disease progression in patients with X-ALD have recently been described (21). The changes are an increase in choline and a decrease in NAA. They occur in areas where subsequent lesion progression is observed, but not in the remainder of the brain. These areas were interpreted as a zone of impending or beginning demyelination (zone D). In the current study, however, we did not compare patients with healthy control subjects and could therefore compare only changes within the patient population. In asymptomatic patients who had no changes at MR imaging over a mean follow-up period of 3.5 years, we assumed the white matter to be normal.
In the current study, the NAWM of the asymptomatic patients had a mean NAA value that was significantly lower than that of the healthy control subjects by 17% (7.89 vs 9.52 mmol/L). It is important to note that this was not accompanied by changes in DT imaging measures. A chronic axonal degenerative process, adrenomyeloneuropathy, is known to manifest in adolescence or adulthood in all patients with X-ALD. Although their white matter is considered to be of normal appearance at MR imaging, many patients with adrenomyeloneuropathy show mild deficits in psychomotor speed and visual memory (22). The spinal cord and peripheral nerves show signs of axonal degeneration and a "dying back" phenomenon, with secondary myelin loss (23). Our finding of low NAA value may be the first sign of metabolic impairment of the axon in patients with X-ALD.
NAA is believed to be of neuronal/axonal origin in mature brain (4), and its decrease in patients with X-ALD is most likely due to either axonal loss or dysfunction. It has been suggested that some NAA may be present in oligodendrocytes (24), so it cannot be ruled out that reductions in NAA are also related in part to changes in oligodendrocyte metabolism that are associated with demyelination. However, the almost complete absence of NAA seen in severe involvement and/or chronic X-ALD lesions (2) appears to be the result of irreversible axonal loss (ie, zone A).
The high choline signal intensity in X-ALD lesions is most likely due to enhanced myelin turnover related to demyelination (25) and accumulation of membrane myelin degradation products. The highest choline levels are seen in zones B and C, where the demyelination is most active, whereas the lowest level of choline is seen in the burned-out center of the lesion (zone A). This biphasic gradient across the lesion is seen in Figure 2.
Creatine emits a composite signal consisting of creatine and phosphocreatine, compounds involved in energy metabolism. In vitro, the creatine content of glial cells is two to four times that of neurons (9). An increase in creatine level seen at the edge of the demyelinating lesion in patients with X-ALD probably reflects gliosis. In the burned-out center of the lesion, all metabolites, which include creatine, decrease, probably reflecting the decrease in cellular density and the relative increase in water content.
In white matter, the diffusion process is believed to be affected by the structural organization of the tissue at a microscopic level, such as myelin density, fiber packing, fiber size, and cell types present. Tissues that have a random microstructure or unrestricted media will have diffusion equal in all directions, or isotropic diffusion. Tissues that have an ordered microstructure will exhibit diffusion that is greater in some directions than in others, or anisotropic diffusion. Such anisotropic diffusion is observed in the myelinated fibers of white matter, where diffusion is greatest parallel to the fibers and lowest perpendicular to them.
The degree to which myelin and axons contribute to restriction and anisotropy of water diffusion within the brain remains a topic of discussion. Previous DT MR imaging studies of cerebral white matter development in human premature and term infants (26,27) demonstrated that in general, the apparent diffusion coefficient decreases while relative anisotropy increases with brain maturation. The most prominent regional difference at term is the increased relative anisotropy in the internal capsule, indicative of high directionality of diffusion, which could be related in part to myelination. Current opinion emphasizes that axonal diameter increases before and during myelination (13). This diameter change could also contribute in an important way to the diminished water diffusion perpendicular to the orientation of the fiber and thereby contribute to the increase in relative anisotropy. In vitro nervous system models (28,29) lend little support to the theory that myelin is a governing factor of anisotropy. A study on the determinants of anisotropic water diffusion in nerves (28) showed a similar degree of anisotropy in myelinated and nonmyelinated nerves.
In the current study, we observed that FA decreased and IADC increased over the zones toward the center of the lesion (zone A). The decrease in mean FA indicates the loss of an ordered structure governing the directionality of water molecule displacement. The increase in mean IADC suggests an increase in free water and a decrease in structures that restrict water diffusion. These findings are in agreement with those in a previous study on DT imaging in X-ALD (3).
Our voxel-by-voxel analysis of different areas in a patient with CCALD suggests that FA decreases together with NAA toward the center of the lesion (zone A). Conversely, IADC increases toward the center as NAA decreases. We confirmed this correlation in a direct comparison of all values, which revealed a strong logarithmic relationship between NAA and FA and an inverse logarithmic relationship between NAA and IADC. Since NAA is known to be predominantly present in axons in white matter, this lends credence to the belief that axonal integrity is an important factor governing restriction and anisotropy of water diffusion within the brain (28). However, the logarithmic relationship illustrates the fact that DT imaging measures are less sensitive than NAA measures and show changes only after significant decreases in NAA. An example is given by our observation of low NAA values in asymptomatic patients with X-ALD who had normal conventional MR and DT imaging findings.
Choline and creatine both increase toward the edge of the lesion. This is consistent with the enhanced membrane turnover associated with demyelination and with the increased cellular density associated with inflammation and gliosis. Both choline and creatine correlate poorly with FA and IADC, a finding that indicates that the membrane turnover and cell accumulation associated with beginning demyelination do not have an effect on diffusional properties in ordered axonal systems. In vivo studies (29) have shown that despite an intact myelin sheath in early wallerian degeneration of peripheral nerve, increases in IADC occur. Within the AWM, the correlation was slightly better, reflecting the fact that with an increase in free extracellular water, all metabolites decrease.
Overall, a strong correlation was observed between NAA levels and FA, reinforcing the concept that both reflect axonal integrity. However, proton MR spectroscopic imaging revealed a low NAA level in regions with normal MR and DT imaging findings. To the best of our knowledge, this observation has not been described before and may be the first sign of axonal impairment in patients with a disease known to manifest as a myeloneuropathy in adulthood. In contrast, DT imaging showed no abnormalities outside the lesion on T2-weighted images. Further, the membrane turnover and cell accumulation associated with beginning demyelination, recognized in the enhanced choline and creatine signal intensity at proton MR spectroscopic imaging, did not have an effect on diffusion parameters. These observations indicate that proton MR spectroscopic imaging has higher sensitivity than both conventional MR and DT imaging for early detection of abnormalities related to demyelination or axonal loss in patients with X-ALD.
Finally, we would like to emphasize that this is a hypotheses-generating work and that on the basis of our results, we have preliminary evidence that requires further exploration and validation. With regard to the concern about the statistically significant findings that result from the number of hypothesis tests performed with an 
level of .05, we expect, by chance alone, that 5% of our hypothesis tests would yield significant P values, even if there were no significant associations present. As such, in analyses 2 and 3, it is possible that some of our significant findings are spurious. Because of this possibility, we plan to perform further studies to confirm our findings.
|
ACKNOWLEDGMENTS |
|---|
|
FOOTNOTES |
|---|
Author contributions: Guarantors of integrity of entire study, H.W.M., P.B.B., E.R.M.; study concepts, F.S.E., P.B.B., E.R.M., H.W.M., G.V.R.; study design, F.S.E., R.I., P.B.B.; literature research, G.V.R., E.R.M., P.B.B.; clinical studies, S.M., R.I., P.C.M.V.Z.; data acquisition, R.I., F.S.E., P.B.B.; data analysis/interpretation, F.S.E., R.I., E.S.G.; statistical analysis, E.S.G.; manuscript preparation, F.S.E., P.B.B.; manuscript definition of intellectual content, F.S.E., P.B.B., E.S.G., P.C.M.V.Z.; manuscript editing, E.S.G., E.R.M., G.V.R.; manuscript revision/review, H.W.M., R.I., S.M., G.V.R., E.R.M.; manuscript final version approval, all authors.
|
REFERENCES |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODS RESULTSDISCUSSIONREFERENCES |
|---|
This article has been cited by other articles:
|
|
 |
|
  A. Bizzi, G. Castelli, M. Bugiani, P.B. Barker, E.H. Herskovits, U. Danesi, A. Erbetta, I. Moroni, L. Farina, and G. Uziel Classification of Childhood White Matter Disorders Using Proton MR Spectroscopic Imaging AJNR Am. J. Neuroradiol., August 1, 2008; 29(7): 1270 - 1275. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 S. Marino, M. De Luca, M. T. Dotti, M. L. Stromillo, P. Formichi, P. Galluzzi, M. Mondelli, P. Bramanti, A. Federico, and N. De Stefano Prominent brain axonal damage and functional reorganization in "pure" adrenomyeloneuropathy Neurology, September 18, 2007; 69(12): 1261 - 1269. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
E. Goebell, J. Fiehler, X.-Q. Ding, S. Paustenbach, S. Nietz, O. Heese, T. Kucinski, C. Hagel, M. Westphal, and H. Zeumer Disarrangement of Fiber Tracts and Decline of Neuronal Density Correlate in Glioma Patients--A Combined Diffusion Tensor Imaging and 1H-MR Spectroscopy Study AJNR Am. J. Neuroradiol., August 1, 2006; 27(7): 1426 - 1431. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 C. S. Cox, P. Dubey, G. V. Raymond, A. Mahmood, A. B. Moser, and H. W. Moser Cognitive Evaluation of Neurologically Asymptomatic Boys With X-linked Adrenoleukodystrophy Arch Neurol, January 1, 2006; 63(1): 69 - 73. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 C. C. Leite, U. C. Reed, M. C. G. Otaduy, M. T. C. Lacerda, M. O. R. Costa, L. G. Ferreira, M. S. Carvalho, M. B. D. Resende, S. K. N. Marie, and G. G. Cerri Congenital Muscular Dystrophy with Merosin Deficiency: 1H MR Spectroscopy and Diffusion-weighted MR Imaging Radiology, April 1, 2005; 235(1): 190 - 196. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
H. W. Moser and P. B. Barker Magnetic resonance spectroscopy: A new guide for the therapy of adrenoleukodystrophy Neurology, February 8, 2005; 64(3): 406 - 407. [Full Text] [PDF]
|
|
|
|
|
|
G. Oz, I. Tkac, L. R. Charnas, I. Y. Choi, K. J. Bjoraker, E. G. Shapiro, and R. Gruetter Assessment of adrenoleukodystrophy lesions by high field MRS in non-sedated pediatric patients Neurology, February 8, 2005; 64(3): 434 - 441. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
P. Dubey, A. Fatemi, P. B. Barker, M. Degaonkar, M. Troeger, K. Zackowski, A. Bastian, S. A. Smith, M. G. Pomper, H. W. Moser, et al. Spectroscopic evidence of cerebral axonopathy in patients with "pure" adrenomyeloneuropathy Neurology, January 25, 2005; 64(2): 304 - 310. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 D. M. Yousem, S.-K. Lee, D. I. Kim, and J. Kim Invited Commentary * Authors' Response RadioGraphics, January 1, 2005; 25(1): 66 - 68. [Full Text] [PDF]
|
|
|
|
|
|
S. S.-F. Peng, W.-Y. I. Tseng, Y.-H. Chien, W.-L. Hwu, and H.-M. Liu Diffusion Tensor Images in Children with Early-Treated, Chronic, Malignant Phenylketonuric: Correlation with Intelligence Assessment AJNR Am. J. Neuroradiol., October 1, 2004; 25(9): 1569 - 1574. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 P. B. Barker and A. Horska Topical Review: Neuroimaging in Leukodystrophies J Child Neurol, August 1, 2004; 19(8): 559 - 570. [Abstract] [PDF]
|
|
|
|
|
|
A. Fatemi, P.B. Barker, A.M. Ulug, L.M. Nagae-Poetscher, N.J. Beauchamp, A.B. Moser, G.V. Raymond, H.W. Moser, and S. Naidu MRI and proton MRSI in women heterozygous for X-linked adrenoleukodystrophy Neurology, April 22, 2003; 60(8): 1301 - 1307. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| RADIOLOGY | RADIOGRAPHICS | RSNA JOURNALS ONLINE |
