Radiology
HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS
QUICK SEARCH:   [advanced]


     

DOI: 10.1148/radiol.2293021462
This Article
Right arrow Abstract Freely available
Right arrow Figures Only
Right arrow Full Text (PDF)
Right arrow Submit a response
Right arrow Alert me when this article is cited
Right arrow Alert me when eLetters are posted
Right arrow Alert me if a correction is posted
Right arrow Citation Map
Services
Right arrow Email this article to a friend
Right arrow Similar articles in this journal
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Right arrow reprints & permissions
Citing Articles
Right arrow Citing Articles via HighWire
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Zhai, G.
Right arrow Articles by Gilmore, J. H.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Zhai, G.
Right arrow Articles by Gilmore, J. H.
(Radiology 2003;229:673-681.)
© RSNA, 2003

Neuroradiology

Comparisons of Regional White Matter Diffusion in Healthy Neonates and Adults Performed with a 3.0-T Head-only MR Imaging Unit1

Guihua Zhai, MS, Weili Lin, PhD, Kathy P. Wilber, BS, Guido Gerig, PhD and John H. Gilmore, MD

1 From the Departments of Biomedical Engineering (G.Z., W.L.), Radiology (W.L., K.P.W.), Neurology (W.L.), Psychiatry (G.G., J.H.G.), and Computer Sciences (G.G.), University of North Carolina at Chapel Hill, CB #7515, Chapel Hill, NC 27599. Received November 14, 2002; revision requested January 15, 2003; final revision received April 24; accepted May 20. Supported in part by center grants HD 03110 and MH 33127 and R01 grant NS 37312. Address correspondence to W.L. (e-mail: linw@email.unc.edu).


   ABSTRACT
TOP
 ABSTRACT
INTRODUCTION
 MATERIALS AND METHODS
RESULTS
 DISCUSSION
 REFERENCES
 
PURPOSE: To evaluate the normal brains of adults and neonates for regional and age-related differences in apparent diffusion coefficient (ADC) and fractional anisotropy (FA).

MATERIALS AND METHODS: Eight healthy adults and 20 healthy neonates were examined with a 3.0-T head-only magnetic resonance (MR) imaging unit by using a single-shot diffusion-tensor sequence. Trace ADC maps, FA maps, directional maps of the putative directions of white matter (WM) tracts, and fiber-tracking maps were obtained. Regions of interest—eight in WM and one in gray matter (GM)—were predefined for the ADC and FA measurements. The Student t test was used to compare FA and ADC between adults and neonates, whereas the Tukey multiple-comparison test was used to compare FA and ADC in different brain regions in the adult and neonate groups.

RESULTS: A global elevation in ADC (P < .001) in both GM and WM and a reduction in FA (P < .001) in WM were observed in neonates as compared with these values in adults. In addition, significant regional variations in FA and ADC were observed in both groups. Regional variations in FA and ADC were less remarkable in adults, whereas neonates had consistently higher FA values and lower ADC values in the central WM as compared with these values in the peripheral WM. Fiber tracking revealed only major WM tracts in the neonates but fibers extending to the peripheral WM in the adults.

CONCLUSION: There were regional differences in FA and ADC values in the neonates; such variations were less remarkable in the adults.

© RSNA, 2003

Index terms: Brain, diffusion • Brain, growth and development • Brain, MR, 13.121412, 13.121413, 13.121416, 13.121417, 13.12144 • Magnetic resonance (MR), diffusion study, 13.12144


   INTRODUCTION
TOP
 ABSTRACT
 INTRODUCTION
MATERIALS AND METHODS
RESULTS
 DISCUSSION
 REFERENCES
 
At diffusion-weighted (DW) magnetic resonance (MR) imaging, signal intensity is related to the relative mobility of endogenous tissue water molecules (1). The signal intensity alterations observed on DW images reflect, in statistical terms, the displacement distribution of the water molecules within voxels. For the water molecules that are not constrained, the direction of motion of a given molecule is random and the molecule’s approximate position over time can be described by using a Gaussian distribution. When the extent of water diffusion in tissues is identical in all directions, the diffusion is normally referred to as isotropic diffusion. Otherwise, the diffusion is referred to as anisotropic diffusion, in which water diffusion is preferentially restricted along one direction. Isotropic diffusion is normally seen in gray matter (GM), whereas anisotropic diffusion is commonly observed in white matter (WM); diffusion in the direction of the fibers is faster than that in the perpendicular direction (2,3).

Although the exact underlying biophysical mechanism(s) responsible for the observed anisotropic behavior in WM is still being extensively investigated, the generally held view is that the specific organization of bundles of more or less myelinated axonal fibers restrictswater molecule motion such that it is preferentially along the direction of the fiber (3–5). Therefore, it has been suggested that DW MR imaging sequences that involve a specific arrangement of diffusion gradients can be used to obtain unique in vivo information about the structural and geometric organization of tissues, especially during early brain development (2,6).

Typically, diffusion gradients oriented in at least six noncolinear directions are used to acquire a set of images, which are used to estimate the diffusion tensor in each voxel (4,5). Subsequently, values that characterize specific features of the diffusion process, such as the principal diffusivities (ie, eigenvalues of the diffusion tensor), the indexes of diffusion anisotropy, and the principal directions of diffusion (ie, eigenvectors of the diffusion tensor), are computed (4,5,7). With use of these parameters, extensive research efforts have been devoted to using diffusion-tensor imaging for both patients and healthy subjects to gain new insights into the microstructural organization of WM that are not possible with conventional MR imaging techniques (8–16).

Diffusion-tensor MR imaging has also been used to study brain development, and consistent results have been reported in the literature. An elevation in apparent diffusion coefficient (ADC) in both GM and WM and a reduction in diffusion anisotropy in WM are normally observed in neonates and children as compared with these measurements in adults (17–25). In addition, ADC has been shown to correlate negatively with age, whereas fractional anisotropy (FA) has been shown to correlate positively with age; these findings suggest potential associations among ADC, FA, and brain development (17–24). However, most of these reported results were obtained in sedated subjects who were scheduled to undergo MR imaging for clinical indications. Even though only subjects who had negative MR imaging findings were included in these data analyses, the results may not necessarily reflect the values for healthy subjects. In addition, results of regional measurements of diffusion anisotropy remain lacking.

The protocols for imaging sedated children may not be useful for research studies of the brain development in healthy children and children at high risk for brain abnormalities, the sedation of whom is not appropriate. Thus, the purpose of our study was to evaluate the normal brains of adults and neonates for regional and age-related differences in ADC and FA.


   MATERIALS AND METHODS
TOP
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
RESULTS
 DISCUSSION
 REFERENCES
 
Subjects
A total of 20 healthy neonates (10 male and 10 female newborns with a mean age of 16 days ± 4 [SD]) and eight healthy adults (four men and four women with a mean age of 28 years ± 9) were recruited for this study. Informed consent was obtained from all subjects or their parents, and the experimental protocols were approved by the institutional review board. Neonates admitted to the newborn nursery of the hospitals at our institution were randomly recruited by a research nurse who visited the nursery 2 days a week. Each newborn was examined by a pediatrician at admission to the nursery and found to be healthy. In contrast, the adult subjects were recruited from the community at large.

Neither the adults nor the neonates were sedated for MR imaging. Before the neonates were imaged, they were fed, swaddled, and fitted with ear protection; and their heads were fixed in a vacuum-fix device (Vac-Lok; Med-Tec, Orange City, Iowa). A pulse oximeter was used to monitor heart rate and oxygen saturation. Most of the neonates slept during the MR imaging examination. Since a head-only MR imaging unit was used in this study, the parents were able to sit near their babies throughout the entire examination.

MR Imaging
All images were acquired by using a head-only 3.0-T MR imaging unit (Allegra; Siemens Medical Systems, Erlangen, Germany) with a maximum gradient strength of 40 mT/m and a maximum slew rate of 400 mT/m/msec. Two main MR imaging sequences were used: a T1-weighted magnetization-prepared rapid gradient-echo (MP-RAGE) sequence and a single-shot echo-planar diffusion-tensor sequence.

For the neonate group, the parameters used for the diffusion-tensor MR imaging sequence were as follows: repetition time msec/echo time msec, 4,219/92.2; section thickness, 5 mm; in-plane resolution, 1.72 x 1.72 mm2 with an intersection gap of 1 mm; 12 signals acquired; and 20 sections. Seven images were acquired for each section: One image was acquired without a diffusion gradient (b)—that is, 0 sec/mm2—and the remaining six images were acquired with a b of 1,000 sec/mm2 and diffusion gradients along (1/{surd}2, 0, 1/{surd}2), (-1/{surd}2, 0, 1/{surd}2), (0, 1/{surd}2, 1/{surd}2), (0, 1/{surd}2, -1/{surd}2), (1/{surd}2, 1/{surd}2, 0), and (-1/{surd}2, 1/{surd}2, 0) in (x, y, z) separately.

The parameters used for MP-RAGE MR imaging in the neonates were as follows: 11/4.3/400 (repetition time msec/echo time msec/inversion time msec), 1-mm section thickness, and an in-plane resolution of 0.90 x 0.90 mm2. A total of 122 sagittal images that depicted the entire brain were acquired, and the total data acquisition time was 5 minutes 34 seconds.

For the adult group, all imaging parameters were identical to those used for the neonates, with the exceptions that the inversion time was changed to 300 msec and a total of 128 sections were acquired with the MP-RAGE sequence, resulting in a total data acquisition time of 6 minutes 39 seconds.

Image Analysis
All diffusion-tensor MR images were transferred to a personal computer (Intel Pentium III; Dell, Round Rock, Tex) for data analysis. First, trace images were created by averaging the data from all six diffusion-tensor images. Subsequently, ADC maps were obtained by using the following equation:

where Sb=1,000 is the signal intensity on the trace DW images and Sb=0 is the signal intensity on the image obtained without diffusion gradients. Subsequently, the pixel-by-pixel diffusion tensor was represented by a 3 x 3-matrix diffusion tensor. Analytical expressions were then obtained for the three principal diffusivities of the diffusion tensor, {lambda}1, {lambda}2, and {lambda}3, by solving the characteristic equation for the diffusion tensor. The eigenvectors of the diffusion tensor were also obtained. Subsequently, FA was calculated as follows:

where <{lambda}> = ({lambda}1 + {lambda}2 + {lambda}3)/3. Since FA is a measure of the fraction of the magnitudes of the diffusion tensor that can be ascribed to anisotropic diffusion, it varies between 0 (isotropic diffusion) and 1 (infinite anisotropy). Furthermore, so that a pixel-by-pixel map of fiber directions could be constructed, the putative direction of a WM tract was defined as the eigenvector corresponding to the largest eigenvalue.

To focus only on the major WM tracts (ie, splenium of corpus callosum, genu of corpus callosum, and internal capsule) in the comparison of fiber directions between neonates and adults, FA thresholds of 0.5 and 0.3 were used for the adult and neonate groups, respectively. The choice of threshold values was based on qualitative estimations of FA values in both the neonate and adult groups in an attempt to preserve only the major WM tracts on the fiber directional maps. Therefore, only the pixels with an FA greater than the predefined threshold value were used to calculate the eigenvectors.

Fiber Tracking
The principal diffusive direction of the local tensor—that is, the eigenvector associated with the largest eigenvalue—indicates the local orientation of fiber tracts. Tracing these local directions through WM between user-defined source and target regions enables the reconstruction of three-dimensional trajectories that follow major axonal tracts. Different techniques that are based mostly on path finding in three-dimensional vector fields have been developed (26). A fiber-tracking method originally developed by Melhem et al (27) and Mori et al (28) was modified to include volumes of interest of arbitrary shape as source and target regions for trajectories and was used in our study.

Volumes of interest of the splenium of corpus callosum and the genu of corpus callosum were first defined by using the FA maps with a multiplanar visualization tool. We defined the target as the whole-brain cortex to search for all possible trajectories originating from the two source locations. Subsequently, the resulting traces were overlaid with transverse and midsagittal contour plots of the Sb=0 images to augment three-dimensional visualization. Diffusion-tensor images obtained in three randomly selected neonates and three randomly selected adults were used for fiber tracking to explore the potential discrepancies in fiber tracts between neonates and adults.

Region-of-Interest Analysis
A region-of-interest (ROI) approach was used to measure FA and ADC in the adults and neonates. All ROIs were placed by one of the authors (G.Z.). In each subject, nine ROIs were predefined on a single transverse section through the level of the basal ganglia. Eight of the nine ROIs were placed in WM, including the anterior and posterior limbs of the internal capsule, the genu and splenium of the corpus callosum, and the left and right occipital and frontal WM adjacent to the cortical GM. Since the ROI measurements were performed after images were acquired in all subjects, the image section used for the ROI measurements for each subject was selected at the same time to ensure that all ROIs were placed at similar anatomic locations among the subjects. This approach helped minimize potential experimental errors caused by variability in ROI placements.

Results obtained from the anterior and posterior limbs of the internal capsule, the left and right occipital WM, and the left and right frontal WM were averaged separately for subsequent data analysis. These WM structures were chosen because they exhibited visible anisotropy and were easily identified on diffusion-tensor images. The remaining ROI was placed in the cortical GM. The ROI sizes varied between the adult and neonate groups. For the adult group, the ROI sizes ranged between three and 50 pixels, with the splenium of corpus callosum having the largest ROI (50 pixels) and the cortical GM having the smallest ROI (three pixels). For the neonate group, the ROI sizes ranged between three and 11 pixels, with the splenium of corpus callosum having the largest ROI (11 pixels) and the cortical GM having the smallest ROI (three pixels). To take special care to exclude the cerebrospinal fluid when choosing ROIs, we compared the FA and ADC maps with the Sb=0 images.

Statistical Analyses
The Student t test was use to compare experimentally measured FA and ADC values between the adults and neonates. In addition, a single-factor analysis of variance to correct for multiple comparisons—specifically, the Tukey multiple-comparison test—was used to determine whether regional differences in FA and ADC were significant. P < .05 was considered to indicate significance at a 95% confidence level.


   RESULTS
TOP
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
RESULTS
DISCUSSION
 REFERENCES
 
We were able to obtain high-quality images without apparent motion artifact in 13 of the 20 neonates (eight male, five female). Therefore, the results obtained from the seven neonates whose images exhibited motion artifacts were excluded from the following data analyses. Representative sagittal and reconstructed transverse MP-RAGE MR images from one neonate (Fig 1, A, B) and one adult (Fig 1, C, D) are shown. Although the total data acquisition time for MP-RAGE MR imaging was shorter in the neonate group, the image quality was sufficient. This finding suggests that adequate anatomic MR images can be obtained in neonates without sedation. The GM-WM contrast on the images obtained in the adults was superior to that on the images obtained in the neonates. In addition, the GM had a lower signal intensity than the WM in adults, whereas this contrast was reversed in neonates.



View larger version (185K):
[in this window]
[in a new window]
[Download PPT slide]
  		Figure 1. T1-weighted MP-RAGE MR images obtained in, A, B, a neonate (11/4.3/400) and, C, D, an adult (11/4.3/300). Sagittal images (A and C) and transverse images (B and D) generated from multiplanar reconstruction are shown. The GM-WM contrast on the images obtained in the adult is superior to that on the images obtained in the neonate. The signal-to-noise ratio is also higher on the images obtained in the adult. Nevertheless, results demonstrate that adequate anatomic MR images can be obtained in nonsedated neonates.

 
Typical Sb=0 images (Fig 2, A, D), FA maps (Fig 2, B, E), and ADC maps (Fig 2, C, F) obtained in one neonate (Fig 2, A–C) and one adult (Fig 2, D–F) are shown. Similar to that observed on the T1-weighted MR images, the GM-WM contrast was reversed between the neonate and adult groups on the T2-weighted MR images. In addition, although it was difficult to discern GM from WM on the ADC maps of both the adult group and the neonate group, an obvious difference in FA between the GM and WM was seen for both groups. The FA in the WM was higher than that in the GM. However, only the major WM tracts exhibited higher FA than the GM in the neonate group, as opposed to the whole-brain WM in the adult group, which demonstrated higher FA than the GM.



View larger version (98K):
[in this window]
[in a new window]
[Download PPT slide]
  		Figure 2. Comparisons of, A, D, transverse T2-weighted Sb=0 MR images (4,219/92.2); B, E, FA maps; and, C, F, ADC maps obtained in a neonate (top) and adult (bottom). A, D, Similar to the diffusion-tensor images in Figure 1, these images show the GM to have lower signal intensity than the WM in adults but higher signal intensity than the WM in neonates. B, E, FA maps show the major WM tracts to have higher FA than the GM in both the neonate and the adult. C, F, In contrast, a more homogeneous ADC across the entire brain is seen in both the neonate and the adult.

 
In all ROIs assessed, the ADCs in the neonate group were significantly higher (P < .001) than those in the adult group (Fig 3a). Quantitative measurements in the neonate group revealed mean ADCs (in 10-5 mm2/sec) in WM ROIs ranging between 153.9 ± 10.3 (SD) (in the occipital WM) and 115.0 ± 20.3 (in the splenium of corpus callosum) and a mean GM ROI value of 134.5 ± 22.9. These data indicate that in the neonates, there were significantly higher ADCs in the peripheral WM and the GM than in the central WM, including the corpus callosum and the internal capsule (Table). In contrast, in the adults the ADCs in the WM regions were more homogeneous—mean values ranged between 71.5 ± 4.2 (in the internal capsule) and 88.7 ± 6.7 (in the occipital WM)—with small but significant differences (Table).



View larger version (19K):
[in this window]
[in a new window]
[Download PPT slide]
  		Figure 3a. (a) Graph shows regional ADC measurements in adults (white bars) and neonates (black bars). The SD lines at the top of the bars indicate intersubject variability. In the neonate group, an apparent elevation in ADC is observed for all ROIs evaluated as compared with the ADCs in the adult group. In addition, in the neonate group, ADCs measured in the peripheral WM and the GM are significantly higher than those measured in the major WM tracts. However, this demarcation is less obvious in the adult group. (b) Graph shows a comparison of regional ratios of neonate to adult ADC. A higher ratio is observed in the peripheral WM than in the major WM tracts. frontal = frontal WM, genu = genu of corpus callosum, internal cap = internal capsule, occipital = occipital WM, splenium = splenium of corpus callosum.

 


View larger version (17K):
[in this window]
[in a new window]
[Download PPT slide]
  		Figure 3b. (a) Graph shows regional ADC measurements in adults (white bars) and neonates (black bars). The SD lines at the top of the bars indicate intersubject variability. In the neonate group, an apparent elevation in ADC is observed for all ROIs evaluated as compared with the ADCs in the adult group. In addition, in the neonate group, ADCs measured in the peripheral WM and the GM are significantly higher than those measured in the major WM tracts. However, this demarcation is less obvious in the adult group. (b) Graph shows a comparison of regional ratios of neonate to adult ADC. A higher ratio is observed in the peripheral WM than in the major WM tracts. frontal = frontal WM, genu = genu of corpus callosum, internal cap = internal capsule, occipital = occipital WM, splenium = splenium of corpus callosum.

 

View this table:
[in this window]
[in a new window]

  		P Values for Statistical Comparisons of Regional ADC and FA Values in Adults and Neonates

 
When the ratios of neonate to adult ADC values were compared (Fig 3b), the frontal WM exhibited the highest ratio (2.05); the cortical GM, occipital WM, genu of corpus callosum, internal capsule, and splenium of corpus callosum followed.

In all regions assessed, FA was significantly higher (P < .001) in adults than in neonates (Fig 4a). In the neonate group, quantitative measures of FA revealed mean values ranging between 0.20 ± 0.07 (in the frontal WM) and 0.63 ± 0.06 (in the splenium of corpus callosum) among the WM ROIs and a mean value of 0.10 ± 0.03 in the GM. In contrast, FA varied less remarkably among the WM ROIs in the adults, but there were significant differences in FA between some ROIs (Table).



View larger version (19K):
[in this window]
[in a new window]
[Download PPT slide]
  		Figure 4a. (a) Graph shows regional measurements of FA in adults (white bars) and neonates (black bars). The SD lines at the top of the bars indicate intersubject variability. In addition to lower FA in all ROIs evaluated in the neonate group as compared with the FA measured in the adult group, substantial regional variation in FA in the neonate group also was observed, with the splenium of corpus callosum (splenium) having the highest FA and the GM having the lowest. In contrast, in the adult group, comparable FA was observed in all ROIs evaluated except the GM. (b) Graph shows a comparison of regional ratios of neonate to adult FA. The regional variation in FA observed in the neonate group was also reflected in the ratios of neonate to adult FA. frontal = frontal WM, genu = genu of corpus callosum, internal cap = internal capsule, occipital = occipital WM.

 


View larger version (15K):
[in this window]
[in a new window]
[Download PPT slide]
  		Figure 4b. (a) Graph shows regional measurements of FA in adults (white bars) and neonates (black bars). The SD lines at the top of the bars indicate intersubject variability. In addition to lower FA in all ROIs evaluated in the neonate group as compared with the FA measured in the adult group, substantial regional variation in FA in the neonate group also was observed, with the splenium of corpus callosum (splenium) having the highest FA and the GM having the lowest. In contrast, in the adult group, comparable FA was observed in all ROIs evaluated except the GM. (b) Graph shows a comparison of regional ratios of neonate to adult FA. The regional variation in FA observed in the neonate group was also reflected in the ratios of neonate to adult FA. frontal = frontal WM, genu = genu of corpus callosum, internal cap = internal capsule, occipital = occipital WM.

 
Similar to findings regarding the ratios of neonate to adult ADC values, a substantial regional variation in FA ratios was observed (Fig 4b). However, the order of magnitude of the FA ratios, with values in the GM excluded, was the exact opposite of that observed for the ADC ratios: The splenium of corpus callosum had the highest ratio, and the internal capsule, genu of corpus callosum, occipital WM, and frontal WM followed.

The primary eigenvectors, reflecting the putative directions of the WM fibers, are shown for one neonate (Fig 5, A) and one adult (Fig 5, B). No apparent differences in the putative directions of the WM were observed between the neonate and adult groups; this finding was consistent among all subjects.



View larger version (75K):
[in this window]
[in a new window]
[Download PPT slide]
  		Figure 5. Transverse putative directional diffusion-tensor images (4,219/92.2) of the major WM tracts obtained in, A, a neonate and, B, an adult. The putative directions of the fibers are color coded as follows: Red indicates the left-to-right direction; green, the anteroposterior direction; and blue, the inferosuperior direction. The neonate and adult have similar fiber directions in the major WM tracts, including the corpus callosum and the internal capsule.

 
Our preliminary experience with the described fiber-tracking approaches was encouraging, and the findings were consistent among all subjects evaluated. The major WM tracts could be fully traced from the splenium and genu of the corpus callosum to the cortical areas in the adults (Fig 6, bottom row), whereas only the major WM tracts could be traced in the neonates (Fig 6, top row). All of the tracked tracts are illustrated in black.



View larger version (43K):
[in this window]
[in a new window]
[Download PPT slide]
  		Figure 6. Drawings illustrate fiber-tracking results for a neonate (top drawings) and an adult (bottom drawings). All of the tracked tracts appear in bold black lines. Fibers were tracked from the splenium and genu of the corpus callosum to the brain surface. The drawings illustrate top (left), left (middle), and top-oblique (right) views of the brain. The resulting traces are overlaid with transverse and midsagittal contour plots of the Sb=0 images. It is evident that in the adult group, the WM tracts can be tracked from the splenium and genu of corpus callosum to the cortical areas, whereas in the neonate group, only the WM tracts can be tracked.

 

   DISCUSSION
TOP
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
RESULTS
 DISCUSSION
REFERENCES
 
The results of this study indicate that it is possible to obtain high-quality diffusion-tensor images in nonsedated newborns by using a fast diffusion-tensor sequence with a 3.0-T MR imaging unit. We found that neonates had significantly higher WM ADC and significantly lower WM FA compared with these measurements in adults; these findings are consistent with previous reports (13,17,18,21,22,25). We also observed a significant difference in FA and ADC between the central WM and peripheral WM regions in neonates.

Extensive efforts have been devoted to the investigation of brain development with MR imaging. Signal intensity alterations on conventional T1-weighted and/or T2-weighted MR images are commonly used to facilitate the depiction of brain developments. Although the extents of hyperintensity and hypointensity of the WM seen on T1- and T2-weighted MR images, respectively, have been shown to correlate well with WM development, the sensitivity of conventional MR imaging sequences remains lacking (29). Within the past decade, many investigators have demonstrated that diffusion-tensor imaging is highly sensitive in revealing microstructural changes in WM, with findings that potentially reflect the integrity of myelination and/or the maturity of myelin in neonates (17–25). In most prior studies, the data retrospectively selected for analysis have been from MR imaging studies that were performed in sedated infants for clinical reasons and in which no anatomic abnormalities were found (17–25). These factors raise the question of whether the results reflect true normative values of ADC and FA.

In this study, however, we prospectively recruited healthy full-term neonates. In addition, the use of MR imaging in nonsedated subjects minimized the potential confounding factors in measuring ADC and FA that are associated with sedation. Therefore, the approach used in this study has important clinical and research implications, particularly in the early detection of WM abnormalities in neonates who are at high risk of having neurodevelopmental disorders, which cannot be fully characterized with conventional MR imaging.

Engelbrecht et al (17) performed DW imaging in 44 children ranging in age from 7.0 days to 7.5 years. They observed elevated ADCs for the children compared with values reported for adults. Of the 44 subjects examined, five were about 1 month old, similar to the ages of the neonates examined in our study. For these five subjects, the reported ADCs (at 10-5 mm2/sec) ranged between 104 and 164; these values are in agreement with the ranges of ADCs observed in our study. Engelbrecht et al (17) also reported that among the WM ROIs evaluated, the occipital WM exhibited the highest ADC and the internal capsule the lowest. Nevertheless, the ADCs measured in the occipital WM and internal capsule in their study appear to be higher than those measured in these areas in our study, probably because of the differences in age between the subjects in the two studies.

It has been suggested that the increased water content in the brain of neonates may account for the observed elevation in ADC in neonates compared with the ADCs in adults. Neil et al (22) imaged 22 newborns 36 hours after birth. ADC and diffusion anisotropy were measured in all subjects. Neil et al reported that a correlation existed between ADC and gestational age, suggesting the potential role of brain water in the elevation in ADC in neonates. However, no clear relationship was observed when diffusion anisotropy was correlated with gestational age.

The findings of Neil et al (22) are perhaps not surprising, given the fact that ADC measurement represents an assessment of overall water mobility. In the brains of neonates, the water content is higher than that in the brains of adults and thus potentially leads to less water restriction by either membranes or other physiologic barriers and consequently a higher ADC. In contrast, diffusion anisotropy is more directly associated with the degree of WM myelination and thus should not be affected by the extent of brain water content. Therefore, the potential correlation between age and extent of WM anisotropy may be explained by the status of WM myelination during brain development.

A direct correlation between WM anisotropy and age has been reported by many investigators (17,20,23,30,31). The general finding is that diffusion anisotropy in WM increases between infancy and adulthood and subsequently decreases with age (30,31), most likely reflecting the status of WM myelination. However, the exact relationship between the two parameters appears to vary among different reports: Schmithorst et al (23), who examined subjects aged 5–18 years, concluded that a linear relationship between FA and age existed. In contrast, Mukherjee et al (20) observed an exponential increase in diffusion anisotropy with age in the posterior limb of the internal capsule and in the thalamus of subjects aged 1 day to 11 years, although a linear relationship was also observed in the lentiform nucleus.

Furthermore, the results based on findings in subjects aged 7.0 days to 7.5 years reported by Engelbrecht et al (17) appear to be best characterized by the equation A[1 - exp(-x)], where A is a constant and x is subject age. McGraw et al (25) report similar results for subjects aged 4 days to 71 months. Although the differences in age of the subjects examined could have accounted for the observed discrepancies in the relationship between diffusion anisotropy and age for results reported in the literature, one of the most likely explanations is the potential regional variations in diffusion anisotropy measurements. Substantially and statistically significant regional variations in ADC and FA measurements were found for both the adult and neonate groups in our study. Therefore, in addition to the potential age dependence of FA and ADC, regional variations in both parameters should be considered when comparing results obtained from different studies.

With regard to correlations between regional variations in FA and ADC and myelination in neonates, McGraw et al (25) recently reported regional variation in FA in pediatric subjects aged between 4 days and 71 months. Their study results indicate that the compact WM (ie, major WM tracts) exhibited higher FA than did the noncompact WM (ie, peripheral WM). In good agreement with their reported results, significant regional variations in FA were also observed in our study. Although McGraw et al (25) measured FA in anatomic locations similar to those chosen in our study, most of their data analyses were focused on the comparison between compact and noncompact WM rather than on subregion analysis, as was performed in our study.

Among the WM ROIs evaluated in our study, the splenium of corpus callosum exhibited the highest FA; the genu of corpus callosum, internal capsule, occipital WM, and frontal WM followed. In addition, when the FA of each ROI in the neonate group was normalized to the corresponding region of the adult group, the FA in the splenium of corpus callosum was the most comparable between the neonate and adult groups, whereas the frontal WM was less similar between the groups. In other words, if one hypothesized that FA measurements reflect the degree of myelination, then the splenium of corpus callosum would have the most advanced myelination and the frontal WM would be the least myelinated region 2 weeks after birth.

In contrast, when the ratios of neonate to adult ADC were examined, the order of magnitude was exactly opposite of that observed with FA ratios: The frontal WM exhibited the highest ratio, and the splenium of corpus callosum the lowest. This finding is of interest since it is well known that brain water is expected to decrease as the brain develops, and this has been suggested as one of the primary factors contributing to the elevated ADC in neonates (22). The observed order of ADC ratios implies that the frontal WM is the least advanced and the splenium of corpus callosum the most advanced in myelination; these conclusions are identical to those relating to the FA ratios.

The regional variations in FA and ADC observed in our study are consistent with the known temporal order of brain myelination. As summarized by Volpe (32), the pattern of brain myelination starts from proximal to distal pathways, from sensory to motor pathways, and from projection to associative pathways. Volpe’s (32) study findings also indicate that the occipital lobe completes the myelination process before the frontal lobe completes the process. Therefore, these findings support our hypothesis that regional discrepancies in FA and ADC may indicate different degrees of myelination in different brain regions.

However, one must be cautious in interpreting our findings, given the potential experimental errors induced by variations in ROI placements among subjects and the limited number of subjects examined in our study. More studies, including those involving the longitudinal follow-up of the same subjects, are needed to further assess the implications of the regional variations in FA and ADC that we observed in the neonate group.

A major factor of our study was that all images were acquired with a 3.0-T head-only MR imaging unit, whereas most of the studies were conducted at 1.5 T. With the improved signal-to-noise ratio, it was possible to shorten the total data acquisition time while maintaining an adequate signal-to-noise ratio and thus minimize motion artifacts.

Conversely, our study had two technical limitations. First, the use of a single-shot echo-planar sequence resulted in limited spatial resolution, particularly for the neonate group. This limited spatial resolution probably induced partial volume effects and possibly confounded the ROI measurements of FA and ADC. Second, the results of fiber tracking are preliminary, and, thus, the accuracy of this method has yet to be established. Nevertheless, our preliminary study results suggest that different patterns of fiber-tracked tracts exist between neonates and adults.

In summary, we observed substantial regional variations in FA and ADC in neonates. These variations most likely reflected regional differences in the extent of WM myelination. In addition, our study results represent the true normative values for ADC and FA in healthy neonates.


   ACKNOWLEDGMENTS
 
We are grateful to Christos Davatzikos, PhD, of the University of Pennsylvania, Philadelphia, for providing an early version of the fiber-tracking tool. We are also grateful to Pierre Fillard, of the University of North Carolina at Chapel Hill and Lyon Technical High School of Electronics and Computer Science, Lyon, France, for implementing the code for diffusion-tensor image processing and fiber tracking.


   FOOTNOTES
 
Abbreviations: ADC = apparent diffusion coefficient, DW = diffusion weighted, FA = fractional anisotropy, GM = gray matter, MP-RAGE = magnetization-prepared rapid gradient echo, ROI = region of interest, WM = white matter

Author contributions: Guarantors of integrity of entire study, W.L., J.H.G., G.G.; study concepts, G.Z., W.L., J.H.G., G.G.; study design, all authors; literature research, G.Z., W.L., J.H.G., G.G.; clinical studies, J.H.G.; experimental studies, all authors; data acquisition, K.P.W., W.L., J.H.G.; data analysis/interpretation, all authors; statistical analysis, G.Z., W.L.; manuscript preparation and definition of intellectual content, all authors; manuscript editing and revision/review, G.Z., W.L., J.H.G., G.G.; manuscript final version approval, all authors


   REFERENCES
TOP
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
RESULTS
 DISCUSSION
 REFERENCES
 

  1. Tsuruda JS, Chew WM, Moseley ME, Norman D. Diffusion-weighted MR imaging of the brain: value of differentiating between extraaxial cysts and epidermoid tumors. AJNR Am J Neuroradiol 1990; 11:925-931.[Abstract]
  2. Basser PJ. Inferring microstructural features and the physiological state of tissues from diffusion-weighted images. NMR Biomed 1995; 8:333-344.[Medline]
  3. Barkovich AJ. Concepts of myelin and myelination in neuroradiology. AJNR Am J Neuroradiol 2000; 21:1099-1109.[Free Full Text]
  4. Basser PJ, Pierpaoli C. A simplified method to measure the diffusion tensor from seven MR images. Magn Reson Med 1998; 39:928-934.[Medline]
  5. Shrager RI, Basser PJ. Anisotropically weighted MRI. Magn Reson Med 1998; 40:160-165.[Medline]
  6. Le Bihan D. Molecular diffusion, tissue microdynamics and microstructure. NMR Biomed 1995; 8:375-386.[Medline]
  7. Basser PJ, Pierpaoli C. Microstructural and physiological features of tissues elucidated by quantitative-diffusion-tensor MRI. J Magn Reson B 1996; 111:209-219.[CrossRef][Medline]
  8. Ciccarelli O, Werring DJ, Wheeler-Kingshott CA, et al. Investigation of MS normal-appearing brain using diffusion tensor MRI with clinical correlations. Neurology 2001; 56:926-933.[Abstract/Free Full Text]
  9. Griffin CM, Chard DT, Ciccarelli O, et al. Diffusion tensor imaging in early relapsing-remitting multiple sclerosis. Mult Scler 2001; 7:290-297.[Abstract/Free Full Text]
  10. Mainero C, De Stefano N, Iannucci G, et al. Correlates of MS disability assessed in vivo using aggregates of MR quantities. Neurology 2001; 56:1331-1334.[Abstract/Free Full Text]
  11. Rose SE, Chen F, Chalk JB, et al. Loss of connectivity in Alzheimer’s disease: an evaluation of WM tract integrity with colour coded MR diffusion tensor imaging. J Neurol Neurosurg Psychiatry 2000; 69:528-530.[Abstract/Free Full Text]
  12. Wolf RL, Zimmerman RA, Clancy R, Haselgrove JH. Quantitative apparent diffusion coefficient measurements in term neonates for early detection of hypoxic-ischemic brain injury: initial experience. Radiology 2001; 218:825-833.[Abstract/Free Full Text]
  13. McKinstry RC, Miller JH, Snyder AZ, et al. A prospective, longitudinal diffusion tensor imaging study of brain injury in newborns. Neurology 2002; 59:824-833.
  14. Arfanakis K, Haughton VM, Carew JD, Rogers BP, Dempsey RJ, Meyerand ME. Diffusion tensor MR imaging in diffuse axonal injury. AJNR Am J Neuroradiol 2002; 23:794-802.[Abstract/Free Full Text]
  15. Taber KH, Pierpaoli C, Rose SE, et al. The future for diffusion tensor imaging in neuropsychiatry. J Neuropsychiatry Clin Neurosci 2002; 14:1-5.[Free Full Text]
  16. Foong J, Symms MR, Barker GJ, Maier M, Miller DH, Ron MA. Investigating regional WM in schizophrenia using diffusion tensor imaging. Neuroreport 2002; 13:333-336.[CrossRef][Medline]
  17. Engelbrecht V, Scherer A, Rassek M, Witsack HJ, Modder U. Diffusion-weighted MR imaging in the brain in children: findings in the normal brain and in the brain with WM diseases. Radiology 2002; 222:410-418.[Abstract/Free Full Text]
  18. Martin KM, Mustafa MH, Wilkinson ID, et al. Study of pediatric brain development using magnetic resonance imaging of anisotropic diffusion. Magn Reson Med 2002; 48:394-398.[CrossRef][Medline]
  19. Mori S, Itoh R, Zhang J, et al. Diffusion tensor imaging of the developing mouse brain. Magn Reson Med 2001; 46:18-23.[CrossRef][Medline]
  20. Mukherjee P, Miller JH, Shimony JS, et al. Normal brain maturation during childhood: developmental trends characterized with diffusion-tensor MR imaging. Radiology 2001; 221:349-358.[Abstract/Free Full Text]
  21. Mukherjee P, Miller JH, Shimony JS, et al. Diffusion-tensor MR imaging of gray and WM development during normal human brain maturation. AJNR Am J Neuroradiol 2002; 23:1445-1456.[Abstract/Free Full Text]
  22. Neil J, Shiran S, McKinstry R, et al. Normal brain in human newborns: apparent diffusion coefficient and diffusion anisotropy measured by using diffusion tensor MR imaging. Radiology 1998; 209:57-66.[Abstract/Free Full Text]
  23. Schmithorst VJ, Wilke M, Dardzinski BJ, Holland SK. Correlation of WM diffusivity and anisotropy with age during childhood and adolescence: a cross-sectional diffusion-tensor MR imaging study. Radiology 2002; 222:212-218.[Abstract/Free Full Text]
  24. Huppi PS, Maier SE, Peled S, et al. Microstructural development of human newborn cerebral WM assessed in vivo by diffusion tensor magnetic resonance imaging. Pediatr Res 1998; 44:584-590.[Medline]
  25. McGraw P, Liang L, Provenzale JM. Evaluation of normal age-related changes in anisotropy during infancy and childhood as shown by diffusion tensor imaging. AJR Am J Roentgenol 2002; 179:1515-1522.[Abstract/Free Full Text]
  26. Basser PJ, Pajevic S, Pierpaoli C, Duda J, Aldroubi A. In vivo fiber tractography using DT-MRI data. Magn Reson Med 2000; 44:625-632.[CrossRef][Medline]
  27. Melhem ER, Mori S, Mukundan G, Kraut MA, Pomper MG, van Zijl PC. Diffusion tensor MR imaging of the brain and WM tractography. AJR Am J Roentgenol 2002; 178:3-16.[Free Full Text]
  28. Mori S, Kaufmann WE, Davatzikos C, et al. Imaging cortical association tracts in the human brain using diffusion-tensor-based axonal tracking. Magn Reson Med 2002; 47:215-223.[CrossRef][Medline]
  29. Barkovich AJ, Wimberger DM. Magnetic resonance of brain development. In: Kucharczyk J, Moseley M, Barkovich AJ, eds. Magnetic resonance neuroimaging. Boca Raton, Fla: CRC, 1994; 168-203.
  30. Pfefferbaum A, Sullivan EV, Hedehus M, Lim KO, Adalsteinsson E, Moseley M. Age-related decline in brain WM anisotropy measured with spatially corrected echo-planar diffusion tensor imaging. Magn Reson Med 2000; 44:259-268.[CrossRef][Medline]
  31. Abe O, Aoki S, Hayashi N, et al. Normal aging in the central nervous system: quantitative MR diffusion-tensor analysis. Neurobiol Aging 2002; 23:433-441.[CrossRef][Medline]
  32. Volpe JJ. Neurology of the newborn Philadelphia, Pa: Saunders, 1995.



This article has been cited by other articles:


Home page
 Am. J. Roentgenol.Home page
A. M. McKinney, S. A. Kieffer, R. T. Paylor, K. S. SantaCruz, A. Kendi, and L. Lucato
Acute Toxic Leukoencephalopathy: Potential for Reversibility Clinically and on MRI With Diffusion-Weighted and FLAIR Imaging
Am. J. Roentgenol., July 1, 2009; 193(1): 192 - 206.
[Abstract] [Full Text] [PDF]

Home page
 Am. J. Neuroradiol.Home page
M.L. Escolar, M.D. Poe, J.K. Smith, J.H. Gilmore, J. Kurtzberg, W. Lin, and M. Styner
Diffusion Tensor Imaging Detects Abnormalities in the Corticospinal Tracts of Neonates with Infantile Krabbe Disease
AJNR Am. J. Neuroradiol., May 1, 2009; 30(5): 1017 - 1021.
[Abstract] [Full Text] [PDF]

Home page
 Am. J. Neuroradiol.Home page
W. Gao, W. Lin, Y. Chen, G. Gerig, J.K. Smith, V. Jewells, and J.H. Gilmore
Temporal and Spatial Development of Axonal Maturation and Myelination of White Matter in the Developing Brain
AJNR Am. J. Neuroradiol., February 1, 2009; 30(2): 290 - 296.
[Abstract] [Full Text] [PDF]

Home page
 PediatricsHome page
A. Murakami, M. Morimoto, K. Yamada, O. Kizu, A. Nishimura, T. Nishimura, and T. Sugimoto
Fiber-Tracking Techniques Can Predict the Degree of Neurologic Impairment for Periventricular Leukomalacia
Pediatrics, September 1, 2008; 122(3): 500 - 506.
[Abstract] [Full Text] [PDF]

Home page
 Am. J. Neuroradiol.Home page
J.H. Gilmore, W. Lin, I. Corouge, Y.S.K. Vetsa, J.K. Smith, C. Kang, H. Gu, R.M. Hamer, J.A. Lieberman, and G. Gerig
Early Postnatal Development of Corpus Callosum and Corticospinal White Matter Assessed with Quantitative Tractography
AJNR Am. J. Neuroradiol., October 1, 2007; 28(9): 1789 - 1795.
[Abstract] [Full Text] [PDF]

Home page
 Cereb CortexHome page
J. Upadhyay, M. Ducros, T. A. Knaus, K. A. Lindgren, A. Silver, H. Tager-Flusberg, and D.-S. Kim
Function and Connectivity in Human Primary Auditory Cortex: A Combined fMRI and DTI Study at 3 Tesla
Cereb Cortex, October 1, 2007; 17(10): 2420 - 2432.
[Abstract] [Full Text] [PDF]

Home page
 Am. J. Roentgenol.Home page
J. M. Provenzale, L. Liang, D. DeLong, and L. E. White
Diffusion Tensor Imaging Assessment of Brain White Matter Maturation During the First Postnatal Year
Am. J. Roentgenol., August 1, 2007; 189(2): 476 - 486.
[Abstract] [Full Text] [PDF]

Home page
 Am. J. Neuroradiol.Home page
H. Oouchi, K. Yamada, K. Sakai, O. Kizu, T. Kubota, H. Ito, and T. Nishimura
Diffusion Anisotropy Measurement of Brain White Matter Is Affected by Voxel Size: Underestimation Occurs in Areas with Crossing Fibers
AJNR Am. J. Neuroradiol., June 1, 2007; 28(6): 1102 - 1106.
[Abstract] [Full Text] [PDF]

Home page
 RadiologyHome page
C. van Pul, J. Buijs, A. Vilanova, F. G. Roos, and P. F. F. Wijn
Infants with Perinatal Hypoxic Ischemia: Feasibility of Fiber Tracking at Birth and 3 Months
Radiology, July 1, 2006; 240(1): 203 - 214.
[Abstract] [Full Text] [PDF]

Home page
 RadiologyHome page
T. Okada, Y. Miki, Y. Fushimi, T. Hanakawa, M. Kanagaki, A. Yamamoto, S.-i. Urayama, H. Fukuyama, M. Hiraoka, and K. Togashi
Diffusion-Tensor Fiber Tractography: Intraindividual Comparison of 3.0-T and 1.5-T MR Imaging
Radiology, February 1, 2006; 238(2): 668 - 678.
[Abstract] [Full Text] [PDF]

Home page
 Am. J. Neuroradiol.Home page
A. J. Fabiano, M. A. Horsfield, and R. Bakshi
Interhemispheric Asymmetry of Brain Diffusivity in Normal Individuals: A Diffusion-Weighted MR Imaging Study
AJNR Am. J. Neuroradiol., May 1, 2005; 26(5): 1089 - 1094.
[Abstract] [Full Text] [PDF]

This Article
Right arrow Abstract Freely available
Right arrow Figures Only
Right arrow Full Text (PDF)
Right arrow Submit a response
Right arrow Alert me when this article is cited
Right arrow Alert me when eLetters are posted
Right arrow Alert me if a correction is posted
Right arrow Citation Map
Services
Right arrow Email this article to a friend
Right arrow Similar articles in this journal
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Right arrow reprints & permissions
Citing Articles
Right arrow Citing Articles via HighWire
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Zhai, G.
Right arrow Articles by Gilmore, J. H.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Zhai, G.
Right arrow Articles by Gilmore, J. H.

HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS
RADIOLOGY RADIOGRAPHICS RSNA JOURNALS ONLINE