| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
Pediatric Imaging |
, PhD
1 From the Departments of Radiology (C.G.F., A.J.T., R.W., D.D.M.L., L.A.H., A.M.U.) and Pediatrics (A.M.P.), New York Presbyterian Hospital-Weill Medical College of Cornell University, NY; and Department of Radiology, Johns Hopkins Medical Institutions, Baltimore, Md (D.D.M.L.). Received February 6, 2002; revision requested April 9; final revision received January 9, 2003; accepted January 22. Address correspondence to C.G.F., 4 E Short Road South, Grand Isle, VT 05458 (e-mail: sairaallapeikko@yahoo.com).
 |
ABSTRACT |
|---|
 TOP ABSTRACT INTRODUCTIONMATERIALS AND METHODS RESULTSDISCUSSIONREFERENCES |
|---|
MATERIALS AND METHODS: Twenty pediatric patients who received a diagnosis of developmental delay underwent brain MR examinations, including diffusion-tensor MR imaging. The MR findings in these patients were compared with those in 10 age-matched neurodevelopmentally healthy children. Diffusion constant (Dav) and anisotropy were measured bilaterally in regions of interest in the centrum semiovale, corona radiata, internal capsule, corpus callosum, and subcortical white matter of the frontal and parieto-occipital lobes. By using a one-tailed Student t test in the positive direction for Dav and in the negative direction for anisotropy and P < .05 to indicate a significant difference, the Dav and anisotropy values for children with developmental delay were compared with those for children who were neurodevelopmentally healthy.
RESULTS: The children with developmental delay had significant increases in Dav in all measured structures (P, <.001 to <.03). Significant decreases in anisotropy were detected in all white matter fiber tracts studied (P, <.001 to <.03) except the posterior limb of the internal capsule.
CONCLUSION: In the children with developmental delay, diffusion-tensor MR imaging depicted decreases in anisotropy and increases in Dav in the white matter fiber tracts, which appeared to be normal at conventional MR imaging.
© RSNA, 2003
Index terms: Brain, abnormalities, 13.14, 13.87 • Brain, growth and development • Brain, MR, 13.121411, 13.121413, 13.121416, 13.12144 • Children, central nervous system • Magnetic resonance (MR), diffusion study, 13.12144
|
INTRODUCTION |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODS RESULTSDISCUSSIONREFERENCES |
|---|
MR imaging has been used to examine the normal patterns of brain maturation and myelination (10–21). White matter maturation is best depicted on T1-weighted MR images within the first 6 months of life (22). The increased signal intensity in the white matter at T1-weighted MR imaging initially is due to a shortened T1 relaxation time caused by protein, cholesterol, and glycolipids in the developing myelin (22). White matter maturation is seen on T2-weighted MR images from birth to about 18 months. Decreasing signal intensity on T2-weighted MR images correlates with decreasing water content in the brain within the first 2 years of life (10–22). Previous investigators (14,22) have reported that myelination progresses from an inferior to superior location, from a posterior to anterior location, and from a central to peripheral (ie, centrifugal) location.
Other investigators (23–29) have used diffusion-tensor MR imaging to assess normal myelination patterns in healthy children. These investigators have quantified the diffusion constant (Dav) and anisotropy and used these measurements as neuroimaging markers of normal brain maturation in children (23–29). Dav is a measure of how much restriction to water diffusion is present in brain tissue. Anisotropy is a measure of the orientation of white matter fiber tracts. Although investigators have yet to apply these MR imaging parameters to examine children with developmental delay, there are studies (10,30) in which investigators have attempted to correlate developmental delay with the appearance of white matter at conventional MR imaging.
Diffusion-tensor MR imaging is an emerging noninvasive modality that is often used in research and applied to the study of white matter fiber tracts. In neonates, Dav should be high and anisotropy should be low, given the increased water content in the brain and the fewer myelinated axons as compared with these parameters in adults. As myelination proceeds, Dav should decrease and anisotropy should increase. By the age of 2 years, myelination is nearly complete and Dav and anisotropy values begin to resemble those measured in healthy adults (22–29). Information regarding the magnitude and severity of any observed alterations in Dav and anisotropy measurements may be valuable for understanding the pathophysiologic characteristics of developmental delay and/or assessing the long-term functional outcome of children with this condition. Thus, the purpose of our study was to determine whether diffusion-tensor MR imaging can depict abnormalities in pediatric patients who have a diagnosis of developmental delay but structurally normal brain MR imaging results. We hypothesized that diffusion-tensor MR imaging would depict increases in Dav and decreases in anisotropy in children with developmental delay.
|
MATERIALS AND METHODS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODS RESULTSDISCUSSIONREFERENCES |
|---|
The control subjects (n = 10) had nonfocal neurologic examination results, no psychiatric histories, and no noteworthy medical histories. These were patients who had normal results at all laboratory examinations, including blood and cerebrospinal fluid tests, and thus were considered healthy control subjects. All control subjects had been referred for MR imaging because of headache or migraine (n = 4) or to rule out seizure (n = 6). After medical evaluation, however, none of the six patients who underwent MR imaging to exclude seizure foci was considered to have had a seizure. Thus, these patients were considered to be healthy control subjects for the purposes of this study. At the time of this writing, follow-up MR imaging had not been performed in either the control subjects or the children with developmental delay.
The diagnosis of developmental delay in the children was made by a pediatric neurologist (A.M.P.). Forty-five children were excluded from this study because structural abnormalities were detected at MR imaging or because they met the exclusion criteria. Exclusion criteria included a history of neurocutaneous syndromes, metabolic disorders, and/or genetic syndromes. Children also were not included if they had a history of premature birth, hydrocephalus, in utero maldevelopment, meningoencephalitis, accidental or nonaccidental trauma, lead exposure, visual or auditory impairment, hypoxic-ischemic encephalopathy, periventricular leukomalacia, abnormal white matter development (demyelinating or dysmyelinating disease), and/or seizure disorder. Children with a diagnosis of mental retardation, autism, pervasive developmental delay, or cerebral palsy also were excluded. These exclusions helped to restrict the patient population, as much as possible, to children with isolated developmental delay of an unknown cause and normal-appearing MR image findings on a macroscopic structural level.
Neurodevelopmental evaluation of the patients with developmental delay consisted of a review of the child’s medical and family histories and a structured neurologic examination, including cranial nerve assessment and motor, cerebellar function, and sensory testing. Current school placement and results of the most recent IQ and language tests were reviewed. These patients were referred for MR imaging of the brain to exclude a structural lesion as the cause of the developmental delay. A diagnosis of developmental delay was made if the child failed to reach one or more developmental milestones in terms of speech and language skills, motor skills, behavioral development, or learning. In this study, we examined children who had one of the following types of isolated neurodevelopmental delay: isolated motor delay (five children), isolated expressive language delay (two children), isolated receptive language delay (three children), or isolated language delay including expressive and receptive components (10 children).
MR Imaging and Calculations
All patients were examined with a 1.5-T whole-body MR imaging unit (Echospeed; GE Medical Systems, Milwaukee, Wis) equipped with high-performance gradients and a manufacturer-supplied quadrature head coil. The following conventional sequences were performed: sagittal T1-weighted (300/14 [repetition time msec/echo time msec], one signal acquired), transverse T2-weighted fast spin-echo (3,000/91, one signal acquired), transverse fast fluid-attenuated inversion-recovery (10,002/172, inversion time of 2.2 seconds, one signal acquired), transverse T1-weighted (500/14, one signal acquired), and transverse diffusion-weighted echo-planar (6,000/99–100, one signal acquired, b values of 0 and 1,000 sec/mm2) MR imaging. The transverse sequences usually involved the use of a 5-mm section thickness with an intersection gap of 2.5 mm, a 256 x 192 matrix, the same imaging angle along the orbitomeatal line, and a 22- or 24-cm field of view.
For diffusion-weighted MR image acquisition, a 128 x 128 matrix, a 5-mm section thickness with no intersection gap, and a 22 x 22-cm field of view were used; the total acquisition time was 42 seconds. No patients were given contrast material. Eleven of the 20 children with developmental delay were sedated with chloral hydrate (Pharmaceutical Associates PAO, Greensboro, NC) (dose, 50 mg per kilogram of body weight).
A diffusion-tensor pulse sequence, which can be used to measure diffusion in any arbitrary direction, was then performed. This is a multisection, single-shot spin-echo echo-planar pulse sequence (6,000/100, one signal acquired) involving the use of a 128 x 128 matrix, a 5-mm interleaved data acquisition with 30 sections encompassing the entire brain, and a 22-cm field of view. By using this sequence, we acquired the diffusion-weighted images in seven directions (x, y, z, x + y, x + z, y + z, and x + y + z) with a maximum b value of 820 sec/mm2 per gradient axis. The data acquisition time for this part of the examination was 5–10 minutes, depending on the number of diffusion-weighted images acquired. In most cases, we obtained a total of 22 diffusion-weighted images with differing b values.
On a workstation (Sun Microsystems, Santa Clara, Calif), an orientation-independent diffusion map was calculated for each pixel from the diffusion-weighted images by using the following equation: Dav = Trace/3 = (Dxx + Dyy + Dzz)/3, where x, y, and z indicate the directions of diffusion weighting and the second (ie, duplicate) letter is standard mathematical nomenclature to indicate that this is a second-order tensor. By using a multivariate fitting routine, we calculated six diffusion maps corresponding to the six independent elements of the diffusion tensor (D). From these diffusion maps, we calculated an orientationally invariant average diffusion map and an orientationally invariant anisotropy map by using the anisotropy index (UAsurf). The anisotropy index is calculated by comparing the Dav with the Dsurf, which is a recently established diffusion constant derived from the surface of the diffusion ellipsoid (32,33). The anisotropy index, which has high sensitivity, is defined in terms of diffusion coefficients as follows:
|
|
There is no standard accepted way to measure anisotropy. Different groups have used different anisotropy measures, such as the relative anisotropy index or fractional anisotropy, to describe anisotropy (31,34,35). Study results have shown that different anisotropy measures have different sensitivities in the prediction of tissue anisotropy, and the anisotropy index, UAsurf, is the most sensitive measurement for the detection of anisotropic changes in white matter fiber tracts (32,36,37). Because we were interested in white matter fiber tracts, we chose to use this anisotropy index, UAsurf.
By using voxel sizes of 3–5 mm, Dav and anisotropy values in the following regions of interest were calculated for each patient: subcortical white matter of the frontal-temporal and parieto-occipital lobes, centrum semiovale, genu and splenium of corpus callosum, and anterior and posterior limbs of the internal capsule bilaterally (Figure). Frontal subcortical white matter regions of interest specifically included those in the precentral gyrus (ie, motor strip). The subcortical white matter of the Broca area in the posteroinferior frontal lobe and the Wernicke area in the temporal lobe, as well as the subcortical white matter of the angular gyrus in the temporal lobe—an area critical to language processing—were specifically assessed.
|
|
|
Statistical Analysis
Mean Dav and anisotropy values for the patients with and those without developmental delay were calculated. By using the one-tailed Student t test in the positive direction for Dav and in the negative direction for anisotropy and by using P < .05 to indicate significant differences, we compared the Dav and anisotropy values for the children with developmental delay with those for the children who were neurodevelopmentally healthy.
|
RESULTS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODS RESULTSDISCUSSIONREFERENCES |
|---|
|
|
An increase in Dav was observed in all areas of the frontal-temporal lobe evaluated, which included the subcortical white matter of the precentral gyrus (ie, motor strip) and the areas involved in language processing (ie, Broca area, Wernicke area, and subcortical white matter of angular gyrus).
In the children with developmental delay, there was a diffuse decrease in anisotropy in all white matter fiber tracts studied. Compared with the anisotropy values for the age-matched control subjects, the decrease in anisotropy in the children with developmental delay was significant in the centrum semiovale, corona radiata, anterior limb of the internal capsule, parieto-occipital and frontal subcortical white matter regions, and genu and splenium of the corpus callosum. P values ranged from less than .001 to less than .03. In the posterior limb of the internal capsule, a significant difference was not observed. However, in this structure, there was a trend toward decreased anisotropy values.
The decrease in anisotropy was observed in all areas of the frontal-temporal lobe evaluated, which included the subcortical white matter of the precentral gyrus (ie, motor strip) and the areas involved in language processing (ie, Broca area, Wernicke area, and subcortical white matter of angular gyrus).
|
DISCUSSION |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODS RESULTSDISCUSSIONREFERENCES |
|---|
Normal brain maturation and myelination are associated with reduced water diffusion and increased diffusion anisotropy (23–30). Calculating apparent diffusion coefficient (in terms of Dav) and anisotropy values may yield unique markers of normal white matter maturation (23–30). Thus, alterations in Dav and anisotropy in white matter fiber tracts, as quantified by using diffusion-tensor MR imaging, may be sensitive markers of white matter injury or disruption in children with developmental delay. In preliminary studies (25,27), inves-tigators have begun to collect and publish normative anisotropy and Dav data in children so that these values can be compared with those in children who have abnormal brain MR imaging or clinical findings.
In initial MR imaging investigations (10–12,14–21), an orderly progression of myelin maturation based on the macroscopic appearance of white matter structures on MR images has been documented. As myelin matures, there is decreasing axonal water due to microtubule and microfilament production and decreasing extracellular free water due to myelin production and glial process formation (22,28). With an increase in water bound to the precursor molecules of myelin, there is a consequent decrease in free water and increase in T1 shortening (22,28,29). T1 shortening correlates temporally with an increase in cholesterol and glycolipid levels, which accompanies the formation of myelin from oligodendrocytes (22,28). T2 shortening correlates temporally with tightening of the spiral of myelin around the axon (22,28). The reduction of free water and the preparation for myelination help to explain the patterns of signal intensity changes that have been observed on conventional MR images (22,27,28).
Just prior to the onset of myelination, a marked increase in anisotropy and decrease in brain water content occur in the cerebral white matter; these changes precede those that are visible on T1- and T2-weighted MR images and that have been documented in animal models (15,22,25,39–41). These changes, which are detectable on diffusion-tensor MR images, are thought to be related to the development of axolemmal membranes and transmembrane pumps, increases in fiber diameter, and early wrapping of axons by oligodendroglial processes, all of which restrict water motion across the axon prior to myelin sheath development (22–28,41). Anisotropy increases in the medial portions of the white matter of the centrum semiovale rather than in the posterior-to-anterior portions typically depicted on T1- and T2-weighted MR images (22). Thus, the normal shortening of relaxation time in white matter that occurs in the developing brain and the quantitative anisotropy and diffusion values that one can measure are related to multiple concurrent processes. The decreasing water content in maturing white matter, the changes in the water-to-macromolecule ratio caused by the arrival of the precursors of myelin, and the myelination process itself all have an important role in myelin maturation in healthy children.
In this study, in the children with developmental delay, we observed persistent significant increases in Dav and persistent decreases in anisotropy in almost all of the white matter structures studied, whereas in the neurodevelopmentally healthy children, we observed unremarkable Dav and anisotropy values. Furthermore, the abnormal increases in Dav and decreases in anisotropy were observed in all areas of the brain that were assessed. These findings were not restricted to the white matter fiber tracts that are specifically related to motor skills (ie, subcortical white matter of precentral gyrus) or language processing (ie, subcortical white matter of angular gyrus, Broca area, or Wernicke area). However, such associations might be seen with a larger patient sample size or in a longitudinal study.
The abnormalities that we observed in this study may have been related to a disruption in any of the complex interrelated processes that characterize myelin maturation. During the premyelination period, there is an increase in axonal diameter, changes in the axonal membrane, and an increase in the concentration of microtubule-associated proteins (27). Increased or abnormal amounts of intraaxonal water caused by problems in microtubule or microfilament production can result in increased diffusivity and decreased anisotropy. Increased extracellular water volume, despite normal-appearing macroscopic brain MR image findings, may be due to hypomyelination, poor glial processing, or decreased synaptic density, which lead to persistent increases in Dav and abnormally low anisotropy values.
If the physical restriction of water motion across the hydrophobic myelin membrane is impaired or occurs to a lesser degree than is required due to hypomyelination or decreased numbers of normal neurons, then the anisotropy of water diffusion in the brain will remain decreased or abnormal. Although many researchers postulate that hypomyelination or decreased synaptic density has a role in developmental delay, autopsy studies to prove this histopathologically are rare, because the rate of death among children who have idiopathic developmental delay is low (38).
During the myelination process, microstructural damage to developing white matter fiber tracts may lead to the development of abnormal anisotropy. Failure to develop normal myelin precursors and improper encasement by oligodendroglial cells can lead to the changes in quantitative anisotropy or Dav values that we observed. Normally, axonal diameter increases during myelination, and this diameter change may contribute substantially to the decreased water diffusion perpendicular to the orientation of the fiber and thus to the increase in anisotropy (28). Abnormal axonal growth, which can lead to diminished axonal diameter, can result in persistently abnormal anisotropy.
Anisotropy values are normally lower in the central white matter than in the posterior limb of the internal capsule. This is a consequence of the orientation of the white matter fiber tracts. In the centrum semiovale, the fibers in the central cerebral white matter have a multidirectional orientation as opposed to the tightly packed, unidirectional, and roughly parallel orientation of the white matter fibers in the posterior limb of the internal capsule (28,42). Thus, in children, the macroscopic organization of white matter fiber tracts influences anisotropy; this phenomenon has been previously reported to occur in adults (34,35). Abnormalities in the orientation of white matter fiber tracts can cause the abnormalities observed in this study and are likely to be occult at conventional macroscopic brain imaging.
In our study, we observed diffuse abnormality in the white matter structures studied in all of the children with developmental delay, but none of these children had a diagnosis of global or pervasive developmental delay. It may be that even in cases of mild or nonglobal developmental delay, there is more white matter damage on a microscopic level than is suspected clinically or can be tested clinically. Further study is needed to determine whether children with global or pervasive developmental delay have more extensive and quantitative abnormalities at diffusion-tensor MR imaging.
Children who demonstrate normal brain development have Dav and anisotropy values that resemble those for healthy adults. In this study, all of the children with mild developmental delay had markedly abnormal values compared with the neurodevelopmentally healthy children. Although a significant difference in anisotropy measurements in the posterior limb of the internal capsule was not observed between the two groups, these measurements tended to be decreased from the normal values in the children with developmental delay. It may be that a significant difference would have been seen with an increased number of region-of-interest measurements. A significant difference also may have been demonstrated with a larger patient sample size—that is, with a larger enrollment of children with developmental delay.
The diagnosis of developmental delay is made clinically and is age dependent. Prior to the development of clinical diffusion-tensor MR imaging, investigators thought that serial MR examinations would be needed to differentiate arrested myelination from slow but progressive development (11,43). Likewise, serial diffusion-tensor MR imaging may be needed to make a similar assessment.
Prior to the routine clinical use of diffusion-tensor MR imaging, normative data should be acquired because the anisotropy and Dav values measured in the brain parenchyma are age dependent. In the brain of humans, the most rapid changes in myelination occur during the first 9 months following birth (14,15,19), and the MR signal intensity changes related to myelin maturation are complete by age 18–24 months (14,15). Therefore, we chose to initially examine children with developmental delay who were older than 2 years.
With regard to the anisotropy and Dav values, we did not observe significant differences in terms of regional variation (ie, parieto-occipital lobe vs frontal lobe) or between the right and left hemispheres. Any differences could have been secondary to volume averaging of the gray matter cortex or cerebrospinal fluid with white matter in the regions of interest chosen for diffusion-tensor MR imaging; this was a potential weakness of this study. The abnormalities that we observed in the parieto-occipital and frontal subcortical white matter regions in the children with developmental delay may yield insight into the pathophysiologic features of this condition.
In conclusion, in the children with developmental delay, diffusion-tensor MR imaging depicted anisotropy and Dav abnormalities in multiple white matter structures despite normal-appearing brain MR image findings. Further longitudinal studies are needed to determine whether the abnormalities detected at diffusion-tensor MR imaging can be used as diagnostic imaging markers of developmental delay. The magnitude and severity of abnormal diffusion-tensor MR imaging values, if followed up longitudinally, may correlate with therapeutic or functional outcome and long-term prognosis. In addition, diffusion-tensor MR imaging may represent another method for objective and quantifiable assessment of neurodevelopment, especially as more children with developmental delay are examined by using this emerging modality.
|
ACKNOWLEDGMENTS |
|---|
|
FOOTNOTES |
|---|
Author contributions: Guarantor of integrity of entire study, C.G.F.; study concepts and design, C.G.F., D.D.M.L., R.W., A.M.U.; literature research, C.G.F., D.D.M.L.; clinical studies, L.A.H., A.M.P.; experimental studies, C.G.F., L.A.H., A.M.P.; data acquisition and analysis/interpretation, C.G.F., D.D.M.L., A.J.T., A.M.U.; statistical analysis, C.G.F., D.D.M.L., A.J.T., R.W., A.M.U.; manuscript preparation and definition of intellectual content, C.G.F.; manuscript editing and revision/review, C.G.F., A.J.T., R.W., A.M.U.; manuscript final version approval, C.G.F.
|
REFERENCES |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODS RESULTSDISCUSSIONREFERENCES |
|---|
This article has been cited by other articles:
|
|
 |
|
  J. Skranes, T. R. Vangberg, S. Kulseng, M. S. Indredavik, K. A. I. Evensen, M. Martinussen, A. M. Dale, O. Haraldseth, and A.-M. Brubakk Clinical findings and white matter abnormalities seen on diffusion tensor imaging in adolescents with very low birth weight Brain, March 1, 2007; 130(3): 654 - 666. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 T. D. Warner, M. Behnke, F. D. Eyler, K. Padgett, C. Leonard, W. Hou, C. W. Garvan, I. M. Schmalfuss, and S. J. Blackband Diffusion Tensor Imaging of Frontal White Matter and Executive Functioning in Cocaine-Exposed Children Pediatrics, November 1, 2006; 118(5): 2014 - 2024. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 H J Williams Imaging the child with developmental delay Imaging, December 1, 2004; 16(2): 174 - 185. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 W. R. Kates, C. P. Burnette, B. A. Bessette, B. S. Folley, L. Strunge, E. W. Jabs, and G. D. Pearlson Frontal and Caudate Alterations in Velocardiofacial Syndrome (Deletion at Chromosome 22q11.2) J Child Neurol, May 1, 2004; 19(5): 337 - 342. [Abstract] [PDF]
|
|
|
|
 |
|
 B. P. Wood Are Brains Normal in Children with Developmental Delay? AAP Grand Rounds, February 1, 2004; 11(2): 22 - 23. [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| RADIOLOGY | RADIOGRAPHICS | RSNA JOURNALS ONLINE |
