 |
|
 |
|
 |
 |
|
 |
 |
 |
 |
 |
 |
 |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
a Carman and Ann Adams Department of Pediatrics, Children's Hospital of Michigan, Wayne State University, Detroit, Michigan
b Department of Neurology, Children's Hospital of Michigan, Wayne State University, Detroit, Michigan
c Department of Radiology, Children's Hospital of Michigan, Wayne State University, Detroit, Michigan
d Department of Psychiatry, Children's Hospital of Michigan, Wayne State University, Detroit, Michigan
 |
ABSTRACT |
|---|
 TOP ABSTRACT METHODSRESULTSDISCUSSIONREFERENCES |
|---|
METHODS. Fractional anisotropy and apparent diffusion coefficient for uncinate fasciculus, stria terminalis, fornix, and cingulum were measured in 7 right-handed children (5 girls and 2 boys; mean age: 9.7 ± 2.6 years) with a history of early severe socioemotional deprivation in Eastern European orphanages and compared with similar measurements in 7 right-handed normal children (4 girls and 3 boys; mean age: 10.7 ± 2.8 years).
RESULTS. Neuropsychological assessment of the orphans verified the relatively mild specific cognitive impairment and impulsivity consistent with previous studies of children who were adopted from Romanian orphanages. Fractional anisotropy values in the left uncinate fasciculus were decreased significantly in the early deprivation group compared with control subjects. Apparent diffusion coefficient values for the early deprivation group tended to be greater than that in control subjects in all of the tracts measured, without reaching statistical significance.
CONCLUSION. Our study demonstrates in children who experienced socioemotional deprivation a structural change in the left uncinate fasciculus that partly may underlie the cognitive, socioemotional, and behavioral difficulties that commonly are observed in these children.
Key Words: behavior disorders/problems • social-emotional problems • MRI • early childhood • diffusion tensor imaging
Abbreviations: PET—positron emission tomography • fMRI—functional MRI • DTI—diffusion tensor imaging • FA—fractional anisotropy • ADC—apparent diffusion coefficient • CSF—cerebrospinal fluid
The impact of early, severe, socioemotional deprivation on brain anatomy and function in humans has not been studied adequately largely because of a lack of a suitable, noninvasive method. Animal studies, including those in nonhuman primates, strongly indicate that early postnatal neglect and deprivation even with adequate nutrition may be associated with both short-term and long-term changes in brain function.1, 2 Recent advances in neuroimaging techniques have provided a variety of novel and noninvasive approaches to the study of early socioemotional deprivation in children and have begun to increase our understanding of brain regions that are most vulnerable to the effects of deprivation.
In a study using positron emission tomography (PET) scanning of children who had been subjected to early socioemotional deprivation in Romanian orphanages and subsequently adopted in the United States, we reported decreased glucose metabolism in a number of brain regions that belong to or are associated strongly with the limbic system.3 Specifically, when compared with control groups, the orphans showed glucose hypometabolism bilaterally in the orbital frontal gyrus (Brodmann's area 11), infralimbic prefrontal cortex (Brodmann's area 25), hippocampus/amygdala, lateral temporal cortex (Brodmann's area 20), and the brainstem.3 These brain regions are highly interconnected, and, in animal studies, many of these structures mediate neuroendocrine responses to stress4, 5 and are damaged with prolonged stress.2, 6 On the basis of these observations, as well as our own findings on the PET scans, we suggested that the hypometabolism may indicate dysfunction in brain regions that result from the stress of early severe socioemotional deprivation and may be associated with the cognitive and behavioral deficits that were manifested by the Romanian orphans.3
Although PET imaging of glucose metabolism is highly useful in studying cortical and subcortical gray matter brain regions, it is not at all useful in studying white matter because of the low glucose metabolic rate of white matter. This also is true for functional MRI (fMRI), which is an excellent imaging technique to map various cortical and subcortical regions that are activated in response to various stimuli (eg, to map brain regions that are involved in vision, speech, motor activities). However, little information on how brain regions are interconnected by various white matter pathways or the strength of these connections can be gathered by these techniques. Another limitation of fMRI activation studies is the need for highly cooperative subjects. Because the hypometabolic brain regions that are shown on the PET scans of the Romanian orphans are highly interconnected, the present study attempted to determine whether the pathways that link these regions show aberrant connectivity. For this, we used the MRI techniques of diffusion tensor imaging (DTI) and fiber tractography, which allow quantitative measurements of the properties of white matter tracts during normal and abnormal brain development. These quantitative measures are expressed as fractional anisotropy (FA) and apparent diffusion coefficient (ADC). The strength of DTI is in its ability to differentiate different white matter pathways from each other noninvasively and pinpoint abnormalities that are associated with specific fiber tracts. DTI is obtained clinically in the evaluation of brain ischemia and is being used increasingly to study the normal white matter changes that are associated with development, maturation, and aging.7, 8 The use of DTI also is being extended to study white matter disruption that is associated with trauma, brain tumors, metabolic disorders, diffuse axonal injury, neuropsychiatric disorders, AIDS, Alzheimer's disease, multiple sclerosis, and epilepsy and now is being incorporated into the routine neuroimaging protocol at many institutions.7–9 DTI has been used by researchers to probe further the neural connectivity between fMRI activated cortical regions.7
We focused on 4 major pathways of the limbic system: (1) the uncinate fasciculus, which connects the anterior temporal lobe to the frontal lobe10; (2) the stria terminalis, which connects the amygdala to the hypothalamus11, 12; (3) the fornix, which connects the hippocampus with the septal area12; and (4) the cingulum, which is the white matter bundle within the cingulate gyrus.11, 12 In addition, our analysis included the corticospinal tract, a robust, well-defined motor pathway10 where we did not anticipate any abnormalities, to control for the possibility that differences in the DTI measures within limbic pathways may represent more generalized abnormalities that are related to factors other than social deprivation (eg, malnutrition).
|
METHODS |
|---|
|
TOPABSTRACTMETHODSRESULTSDISCUSSIONREFERENCES |
|---|
The control group consisted of 7 right-handed normal children (4 girls and 3 boys; mean age: 10.7 ± 2.8 years). All of the children underwent an intellectual screening as well as semistructured interview to ensure that all had intellectual functioning within normal limits and to rule out any current or historical medical or psychiatric conditions. None of the 7 was taking medications. This study was approved by the Wayne State University Human Investigation Committee, and written informed consent and assent were obtained before enrollment in the study.
The study required 2 visits. During the first visit, neuropsychological assessment and neurologic screening were performed to ensure that children met all of the criteria for participation in the study; this lasted 4 to 6 hours. During the second visit, an MRI scan that included the DTI sequence (see below) was performed.
Neuropsychological Evaluation
The neuropsychological evaluation for the orphan group consisted of evaluation of global, verbal, and nonverbal intellectual functioning; expressive and receptive vocabulary and language processing; verbal and visual memory; achievement; executive functioning (attention, impulsivity); manual dexterity; and behavioral functioning. The battery included the following measures: Wechsler Intelligence Scales for Children–Third Edition14; Token Test for Children15; Comprehensive Evaluation of Language Functions, Third Edition16; Wide Range Assessment of Memory and Learning17; Wide Range Achievement Test, Third Edition18; Gordon Diagnostic System19; and Grooved Pegboard.20 In addition, the Behavioral Assessment Scales for Children21 was administered to evaluate internalizing and externalizing behavioral problems. The psychometric properties of these measures have been well established, and these measures are widely used with both clinical and research populations.22, 23 The evaluation also included a social-historical interview with the child's adoptive parent(s). Data that were collected in the semistructured interview and neurologic screen included the physical, developmental, and behavioral status of the children at the time of adoption, at 1 year after adoption, and at the time of the present study. Evaluation of the control children included intellectual screening with the Wechsler Abbreviated Scales of Intelligence24 and a semistructured interview to rule out current or historical medical or psychiatric conditions.
MRI Acquisition
MRI scans were performed using a 1.5-T magnet (Siemens Sonata, Erlangen, Germany). To obtain high-resolution anatomic image volume, we performed inversion time–weighted Magnetization Preparation Rapid Acquisition Gradient Echo sequence with echo time of 4 milliseconds, repetition time of 800 milliseconds, TI of 420 milliseconds, 3 averages, and flip angle of 20 degrees. The data were reconstructed into 124 coronal slices (matrix size: 256 x 256; pixel size: 0.75 mm) with 1.5-mm slice thickness and no gap. The total scanning time for this 3-dimensional acquisition was 
7 minutes.
Diffusion tensor images were acquired in axial plane with diffusion sensitization gradients applied in 6 noncollinear directions and a b value of 1000 s/mm2. The same imaging parameters were used to acquire 1 T2-weighted image volume (b = 0 s/mm2). All image volumes were acquired using 8 averages to increase the signal-to-noise ratio and to reduce image artifacts. The echo time was 97 milliseconds, and the repetition time was 6.6 seconds. A set of 40 axial slices of 3-mm thickness was acquired without gap covering the whole brain, including the cerebellum. The other imaging parameters were as follows: field of view = 24 x 24 cm2, matrix size = 128 x 128, and voxel size = 1.8 x 1.8 x 3 mm3. The total scanning time for the DTI acquisition was 7 minutes.
Fiber Tractography
First, the entire DTI raw data set was examined in a slice-wise manner to exclude subject movement during the study. Fiber tracking was performed using DTI Studio software (H. Jiang and S. Mori; Johns Hopkins University, Baltimore, MD; cmrm.med.jhmi.edu) based on Fiber Assignment by Continuous Tracking algorithm25 with an FA threshold of 0.20 (except for fornix and stria terminalis, for which it was 0.15) and angle threshold 60 degrees. Selected source and target areas then were fixed using direct rendering from regions of interest (ROIs) selected over the FA color map. The uncinate fasciculus, cingulum, and corticospinal tracts were delineated according to the methods described previously.26, 27 The demonstration of fornix and stria terminalis in their entire length required 3 ROI operations. The seed ROI was placed ("or" operation, which selects all fibers that go through this region) in an axial slice over the temporal stem followed by a target ROI ("and" operation, which retains all fibers that are generated by the "or" operation and go through the second ROI) in the axial plane at the level of the hypothalamus as shown in Fig 1. The resulting fiber bundle contains both the fornix and the stria terminalis. Using the axial slices of b = 0 image (Fig 1E), the amygdala and the hippocampus were identified. The stria terminalis can be obtained by removing the fornix using an "and" operation over the amygdala, whereas the fornix can be obtained by performing a "not" operation (removes any fibers that go through this region) anterior to the hippocampus to remove the stria terminalis. Once a fiber tract was demonstrated in its entire length, the mean FA and ADC values for the entire fiber tract were calculated using the statistical program incorporated in the DTI Studio software. To assess the reliability of fiber tracking, the same investigator (T.J.E.) repeated measurements for the 5 tracts for the control group, 1 week apart, with fiber tracking parameters held constant. Interrater variability also was measured by a separate investigator, who followed the same methods, with fiber tracking parameters held constant.
|
|
.10) associations for age or gender differences on fiber tract were found, these variables were included in subsequent analyses that involved identified tracts as covariates. Repeated measures analysis of variance or repeated measures analysis of covariance were performed with group as the between-subjects factor; side as the within-subjects factor; and age, gender, and IQ as covariates. Initially, the significance for the 2 main effects (group and side) and the interaction (group x side) were examined. When the overall test was significant (P < .05), individual simple effects tests for each individual fiber tract were examined. To account for the multiple comparisons, we applied Bonferroni correction, and significance level was determined to be P < .005.
|
RESULTS |
|---|
|
TOPABSTRACTMETHODSRESULTSDISCUSSIONREFERENCES |
|---|
|
The descriptive statistics for the 4 targeted limbic fiber tracts of interest and the corticospinal tract are presented in Table 3. FA values were greater in the normal control group in most of the tracts measured. However, the only finding that reached significance was for the uncinate fasciculus (Fig 2). In the overall test, the group x side interaction for the uncinate was significant (P = .036), with the early deprivation group having reduced left FA relative to the right, as compared with normal control subjects, who demonstrated relatively equal FA values in the 2 hemispheres. In addition, subsequent between-groups simple effects tests revealed that the only significant (P < .005 was considered significant in view of multiple comparisons performed) between-group finding was for the left uncinate fasciculus (P = .003). With regard to the ADC measure, values for the early deprivation group tended to be greater than that in control subjects in most of the tracts measured. However, none of these findings reached statistical significance.
|
|
|
DISCUSSION |
|---|
|
TOPABSTRACTMETHODSRESULTSDISCUSSIONREFERENCES |
|---|
The human uncinate fasciculus originates like a fan from the anterior 3 temporal convolutions (Brodmann's areas 20 and 38) and the cortical nuclei of amygdala (Brodmann's areas 28, 34, and 36) to terminate in the frontal lobes in the gyrus rectus (Brodmann's area 11), the medial orbital cortex (Brodmann's area 11), and subcallosal area (Brodmann's area 25).10 The major finding in the present study of reduced FA in the left uncinate fasciculus provides evidence to indicate structural changes of a relevant brain pathway in children who have experienced early severe socioemotional deprivation. This finding is consistent with and extends the results from our previous study,3 which reported glucose hypometabolism in limbic brain regions, including lateral temporal cortex (Brodmann's area 20), the amygdala, and Brodmann's areas 25 and 11 in the frontal lobe, all regions that are related to either the origin or the termination of the uncinate fasciculus. However, the functional changes in glucose metabolism on the PET scans were found bilaterally. The finding of reduced FA in the left uncinate in the early deprivation group reflects structural defects in this pathway and is presumed to impair the function of the neural network that promotes communication between these brain regions; therefore, it is suspected that abnormalities in this neural network and associated dysfunction in the affected brain regions may underlie, at least partly, the cognitive, socioemotional, and behavioral difficulties that commonly are observed in children with early severe socioemotional deprivation.
Reduced FA in the left uncinate fasciculus also has been reported in schizophrenic patients in several studies,33–35 and this finding was associated with poorer declarative-episodic verbal memory, providing support for the notion that reduced integrity of this pathway is associated with disrupted functions that are presumed to be subserved by these regions and neural network.36 Given that the effects of social deprivation are somewhat broad and heterogeneous and the small sample size in the present study, we cannot link definitely the DTI findings to socioemotional and/or neurocognitive variables. However, given that quantitative DTI measures of the left uncinate fasciculus have been correlated with neurocognitive variables such as general intelligence, visual and verbal memory, and executive function in various patient populations35, 36 and that the participants in our study manifested difficulties in neurocognitive and behavioral functioning (eg, relative deficits in verbal memory was a consistent), it may well be the case that abnormalities in the uncinate fasciculus at least in part underlie some of these difficulties. The mechanisms that account for the reduced FA in the left uncinate fasciculus of the children with early severe socioemotional deprivation are not yet clear, but several potential explanations can be suggested. For example, it is possible that as a result of social deprivation, this tract is less well myelinated than other tracts. A second potential explanation is that fiber numbers are reduced, and a third explanation is that the left uncinate fasciculus is disorganized (hence more isotropic). These possibilities could be the result of inadequate stimulation of this pathway during postnatal development. Although the corticospinal tracts showed normal measurements bilaterally on DTI, which would argue against a more generalized type of injury to the brain (eg, malnutrition) as the sole cause of FA abnormalities, we also found that socially deprived children had significantly lower head circumference, lower total brain volume (1149 ± 82 mL vs 1330 ± 107 mL; P = .004), gray matter volume (607 ± 54 mL vs 716 ± 61 mL; P = .004), and white matter volume (284 ± 38 mL vs 333 ± 32 mL; P = .026) than control subjects, indicating that the role of a more generalized insult cannot be ruled out.
The concept of inadequate stimulation of certain anatomic structures or pathways in the brain during "critical" or "sensitive" periods of brain development is by no means novel. In the cat visual cortex, there is a critical period that ranges from 4 weeks postnatally to 3 months, during which time various experimental manipulations (eg, monocular deprivation by suturing 1 eyelid closed) can cause neuroanatomic changes that result in altered connectivity.37 In the primate, Harlow et al38 found that infant monkeys that were raised in isolation from birth manifested severe behavioral maladjustment after a sensitive period of 8 months.
The concept of inadequate stimulation during development may be operational in social deprivation, in which a failure to achieve adequate stimulation during the postnatal period may lead to dysfunction of related brain structures, as indicated by glucose hypometabolism on PET imaging3 and DTI findings in the present study. However, this hypothesis could not be tested in the present study because of the small number of subjects and because a number of other factors may have played a role either in the generation or in modifying many of the neuropsychological and behavioral problems that were seen in these children. Studies in the past have demonstrated the effect of maternal depression on behavioral problems and enhanced susceptibility to mental illness.39 Genetic factors also can modify the effects of stress. Caspi et al40 found that a functional polymorphism in the promoter region of the serotonin transporter (5-HTT) gene may moderate the influence of stressful life events on depression. Individuals with 1 or 2 copies of the short allele of the 5-HTT promoter polymorphism exhibited more depressive symptoms, diagnosable depression, and suicidality in relation to stressful life events than individuals who were homozygous for the long allele. Although subsequent caregiving environment as well as environmental enrichment have been shown to reverse many of the effects of early deprivation with improvement in cognitive function in animal experiments,41, 42 whether these dysfunctional brain regions that are identified in children who experienced early deprivation can recover completely with time and interventions is a topic of great interest to researchers and obviously to the parents who have adopted such children.
|
FOOTNOTES |
|---|
Address correspondence to Harry T. Chugani, MD, Division of Pediatric Neurology/PET Center, Children's Hospital of Michigan, 3901 Beaubien Blvd, Detroit, MI 48201. E-mail:hchugani{at}pet.wayne.edu
The authors have indicated they have no financial relationships relevant to this article to disclose.
|
REFERENCES |
|---|
|
TOPABSTRACTMETHODSRESULTSDISCUSSIONREFERENCES |
|---|
This article has been cited by other articles:
 |
|
 |
|
  Z. Petanjek, M. Judas, I. Kostovic, and H. B.M. Uylings Lifespan Alterations of Basal Dendritic Trees of Pyramidal Neurons in the Human Prefrontal Cortex: A Layer-Specific Pattern Cereb Cortex, April 1, 2008; 18(4): 915 - 929. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 D. Nazer and V. J. Palusci Child Sexual Abuse: Can Anatomy Explain the Presentation? Clinical Pediatrics, February 1, 2008; 47(1): 7 - 14. [Abstract] [PDF]
|
|
|
|
 |
|
 C. A. Nelson III, C. H. Zeanah, N. A. Fox, P. J. Marshall, A. T. Smyke, and D. Guthrie Cognitive Recovery in Socially Deprived Young Children: The Bucharest Early Intervention Project Science, December 21, 2007; 318(5858): 1937 - 1940. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
T. J. Eluvathingal, K. M. Hasan, L. Kramer, J. M. Fletcher, and L. Ewing-Cobbs Quantitative Diffusion Tensor Tractography of Association and Projection Fibers in Normally Developing Children and Adolescents Cereb Cortex, December 1, 2007; 17(12): 2760 - 2768. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 M. Augustyn and B. Zuckerman From mother's mouth to infant's brain Arch. Dis. Child. Fetal Neonatal Ed., March 1, 2007; 92(2): F82 - F82. [Full Text] [PDF]
|
|
|
|
 |
|
 Minerva BMJ, July 15, 2006; 333(7559): 156 - 156. [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
 |
||
|
||
Home | My Pediatrics | Journal Information | Current Issue | Past Issues | Subscriptions & Services | Contact Us Click here for faster international access © 2006 American Academy of Pediatrics. All rights reserved. |
||
|
||
left; green, anterior 