Brain Advance Access published online on October 24, 2008
Brain, doi:10.1093/brain/awn247
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
Structural dissociation of attentional control and memory in adults with and without mild traumatic brain injury
1Sackler Institute for Developmental Psychobiology, Weill Cornell Medical College, New York, NY, 2Department of Radiology and Biomedical Imaging, University of California, San Francisco, CA, 3Brain Trauma Foundation, 4Department of Neurological Surgery, 5Department of Radiology, Weill Cornell Medical College, New York, NY, 6Department of Neurological Surgery, University of California, San Francisco, CA, USA, 7Biological Science Department, Sungkyunkwan University, Suwon, South Korea and 8Department of Psychology and Human Development, Vanderbilt University, Nasville, TN, USA
Correspondence to:
Bruce D. McCandliss, 1300 York Ave, Box 140, New York, NY 10021, USA E-mail: bdm2001{at}med.cornell.edu
 |
Summary |
|---|
 Top Summary IntroductionMethodsResultsDiscussionConclusionFundingReferences |
|---|
Memory and attentional control impairments are the two most common forms of dysfunction following mild traumatic brain injury (TBI) and lead to significant morbidity in patients, yet these functions are thought to be supported by different brain networks. This 3 T magnetic resonance diffusion tensor imaging (DTI) study investigates whether microstructural integrity of white matter, as measured by fractional anisotropy (FA) within a small set of individually localized regions of interest (ROIs), is associated with these cognitive domains in normal adults and adults with mild TBI. Results in a sample of 23 normal controls reveal a significant correlation between attentional control and FA within a ROI in the left hemisphere anterior corona radiata. Furthermore, the controls demonstrate a correlation between memory performance and FA in a ROI placed in the uncinate fasciculus. Next, to examine whether these relationships are found in the pathological ranges of attention, memory and microstructural white matter integrity associated with mild TBI, these analyses were applied to a group of 43 mild TBI patients. Results, which generally demonstrated a wider range of attention, memory and FA scores, replicated the correlation between attentional control and FA in left hemisphere anterior corona radiata, as well as the correlation between memory performance and FA in the uncinate fasciculus. These two sets of brain-behaviour relationships were highly specific, as shown by a lack of correlation between attention and uncinate fasciculus FA and the lack of correlation between memory performance and anterior corona radiata FA. Furthermore, a ‘correlational double dissociation’ was demonstrated to exist between two distinct frontal structures independently associated with attention and memory, respectively, via a series of multiple regression analyses in both normal controls and adults with mild TBI. The results of the multiple regression analyses provide direct evidence that tract-specific variation in microstructural white matter integrity among normal controls and among mild TBI patients can account for much of the variation in performance in specific cognitive domains. More generally, such findings suggest that diffusion anisotropy measurement can be used as a quantitative biomarker for neurocognitive function and dysfunction.
Key Words: diffusion tensor imaging; traumatic brain injury; executive function; executive attention; memory
Abbreviations: ACR, anterior corona radiata; ANT, attention network task; CB, cingulum bundle; CVLT-II, California verbal learning test, second edition; DTI, diffusion tensor imaging; FA, fractional anisotropy; ICC, intraclass correlation coefficient; ILF, inferior longitudinal fasciculus; LDFR, long delay free recall; ROI, region of interest; SDFR, short delay free recall; SLF, superior longitudinal fasciculus; TBI, traumatic brain injury; UF, uncinate fasciculus
.
Received March 12, 2008. Revised September 4, 2008. Accepted September 8, 2008.
|
Introduction |
|---|
|
TopSummaryIntroductionMethodsResultsDiscussionConclusionFundingReferences |
|---|
Sub-systems of executive function such as attention and memory are among the most significant human brain processes impacting overall cognitive function. A variety of approaches have been used to better understand these anatomical and functional components. On one end of the spectrum are neuropsychological focal lesion studies in patient populations linking gross frontal lobe damage to executive dysfunction and focal retrograde and anterograde amnesia (Stuss and Benson, 1984
; Milner et al., 1985; Levine et al., 1998; Stuss and Alexander, 2000). Neuropsychological data clearly highlight the central role of the frontal lobes in executive attention (Stuss et al., 1981; Vendrell et al., 1995). Yet, there are notable limitations of these methods for isolating and dissociating their functional components. Focal lesions are often quite large and variable in spatial extent which limits the precision of findings to gross divisions of the brain. Additionally, focal lesion studies are restricted to studying categorical damage (impaired versus normal) when, in fact, many injuries may exist on a spectrum of severity that may systematically relate to the degree of cognitive impairment.
Functional magnetic resonance imaging (fMRI) studies have also provided insights into sub-systems of executive function. Such studies provide evidence that response conflict in Stroop and Flanker tasks involve the anterior cingulate gyrus (Casey et al., 2000; Gruber et al., 2002; Fan et al., 2005) while working memory processes involve bilateral prefrontal and temporal regions (Landfield, 1988; Cabeza and Nyberg, 2000; Shu et al., 2003). Yet fMRI faces challenges when applied to patients who exhibit deficits on the very tasks required to activate the regions under investigation and is also restricted to studying cortical grey matter.
Magnetic resonance diffusion tensor imaging (DTI) can be used to assess microstructural white matter integrity on a continuous scale and can be used to examine the anatomical connectivity within functional networks (Basser et al., 1994; Basser and Pierpaoli, 1996). Fractional anisotropy (FA), a normalized DTI measure of the degree of directionality of water diffusion, has been shown to be sensitive to white matter microstructural integrity and is known to correlate with a wide range of cognitive skills (Moseley et al., 2002; Medina et al., 2005; Schmithorst et al., 2005; Mabbott et al., 2006; Niogi and McCandliss, 2006). Even after controlling for age-related development, DTI studies have shown that performance in specific cognitive functions is associated with white matter organization and integrity (Catani, 2006; Durston and Casey, 2006; Niogi and McCandliss, 2006).
Such methods may shed new light on functional dissociations between prefrontal brain circuitry associated with attentional control and memory. It is plausible that examining normal FA variation in white matter tracts and associated cognitive function can provide evidence to determine which pathways impact memory and which pathways impact attentional control. Moreover, variation in white matter integrity, whether due to normal development or pathology, should account for differences in cognitive function. A convenient population to examine a larger range of variation in white matter integrity, memory function, and executive function skills, is those affected by mild traumatic brain injury (TBI). Mild TBI is a potential model human system to study the relative contributions of white matter tracts to executive and memory functions, as the mechanism of damage has been linked primarily to diffuse axonal injury (DAI) (Gennarelli and Graham, 1998; Medana and Esiri, 2003; Povlishock and Katz, 2005). Due to the variable nature of head injury, certain tracts may be damaged to different degrees while other tracts may be relatively spared. Furthermore, attention and memory are the two most commonly affected cognitive domains in mild TBI (Lundin et al., 2006; Malojcic et al., 2008). It is likely that the same tracts that are responsible for normal inter-individual differences in memory and attentional control are selectively damaged in mild TBI, producing these chronic cognitive dysfunctions with different degrees of severity.
In this study, we first conduct an exploratory analysis in normal adults to test potential associations between white matter tracts and individual differences in two cognitive domains: memory and attentional control. After determining the two white matter tracts primarily associated with memory and attentional control in normal adults, we test whether these structure–function relationships extend to the damaged range in mild TBI patients. We then examine whether impairment is categorical or whether cognitive dysfunction exists on a continuous spectrum of severity related to the extent of white matter injury. Furthermore, regional heterogeneity of injury in mild TBI provides a means to establish specificity of the role these particular white matter tracts have in cognitive function. We test the specificity of these structure–function relationships in mild TBI subjects by establishing a quantitative double dissociation. Such evidence would clarify the white matter networks required in memory and attentional control which may be subtly and independently impacted in mild TBI, causing chronic disability. Damage to these white matter pathways may be predictive of dysfunction in the corresponding cognitive domain, thus extending the utility of DTI to the diagnosis of cognitive sequelae in mild TBI.
|
Methods |
|---|
|
TopSummaryIntroductionMethodsResultsDiscussionConclusionFundingReferences |
|---|
Participants
Subjects in this study included 43 patients prospectively recruited with mild TBI (28 males, 15 females), defined as having a Glasgow Coma Scale score of 13–15 at the time of presentation to the Emergency Department following head trauma as well as post-traumatic amnesia. All mild TBI subjects were imaged at least 1 month post-injury (mean 16.9 months, range 1–53 months). Exclusion criteria included age <17 years or >65 years, any prior history of TBI, any history of neurological or psychiatric illness including drug or alcohol abuse, psychotropic medications that would affect cognitive testing or any contraindication to MR imaging such as pregnancy or ferromagnetic implants. The average age was 32.4 years (range 17–61 years). Table 1 summarizes demographic information for the TBI patients. Normal adult control subjects in this study included 23 healthy volunteers (17 males, 6 females) with an average age of 29.9 years (range 18–58 years). There was no significant difference in mean age, gender, handedness (only one normal control was left-handed) or level of education between the mild TBI group and the control group. Written informed consent was obtained from all subjects in accordance with protocols approved by the Internal Review Board of Weill Cornell Medical College or the Committee on Human Research at the University of California, San Francisco.
|
MRI and DTI acquisition and analysis
Magnetic resonance imaging was acquired on two 3 Tesla (3T) GE Signa EXCITE scanners (GE Healthcare, Waukesha, WI, USA) each equipped with 8-channel phased-array head coils. Identical MR imaging hardware, pulse sequences and acquisition protocols were used at the two sites. Whole-brain DTI was performed with a multislice single-shot spin echo echoplanar pulse sequence (TE = 63 ms, TR = 14 s) using 55 diffusion-encoding directions, isotropically distributed over the surface of a sphere with electrostatic repulsion, acquired at b = 1000 s/mm2, 1 acquisition with b = 0 s/mm2, 72 interleaved slices of 1.8 mm thickness each with no gap between slices, a 128 x 128 matrix that is zero-filled during reconstruction to 256 x 256 and a field of view (FOV) of 230 mm. Parallel imaging was employed using the Array Spatial Sensitivity Encoding Technique (ASSET) with an acceleration factor of 2. Images were post-processed offline using DTIstudio software (Jiang et al., 2006
) to obtain FA maps, apparent diffusion coefficient (ADC) maps and directionally encoded colour FA maps. The following conventional 3T MR imaging sequences were acquired: (i) axial 3D inversion recovery fast spoiled gradient-recalled echo (FSPGR) T1-weighted images (TE = 1.5 ms, TR = 6.3 ms, TI = 400 ms, flip angle of 15°) with 230 mm FOV, 156 1.0 mm contiguous partitions at a 256 x 256 matrix, (ii) axial T2-weighted fluid-attenuated inversion recovery (FLAIR) images (TE = 126 ms, TR = 10 s, TI = 2200 ms) with 220 mm FOV, 47 to 48 3.0 mm contiguous slices at a 256 x 256 matrix, (iii) axial magnetization prepared gradient echo (MPGR) T2*-weighted images (TE = 15 ms, TR = 500 ms, flip angle of 20°) with 220 x 170 mm2 FOV, 47 to 48 3.0 mm contiguous slices at a 256 x 192 matrix. All conventional MR images were interpreted by attending neuroradiologists certified by the American Board of Radiology. None of the control subjects had any clinically significant abnormal findings on their MR scans of the brain.
Region of interest analysis
To avoid several confounds associated with spatial normalization of white matter tracts, a manual ROI approach was adopted to test specific structures throughout the brain selected a priori. Each ROI was placed in an anatomically identifiable white matter tract on the directionally encoded colour FA images, as illustrated in Fig. 1. The ROI procedure consisted of using standard ellipse-shaped ROIs. Selection and analysis was implemented with software written in Interactive Data Language v6.0 (IDL, ITT Visual Information Solutions, Boulder, CO, USA). Ellipses were prescribed on axial directionally encoded anisotropy maps at the centre of each structure in normal-appearing white matter, as assessed from the conventional MR imaging sequences. The size and dimensions of the ellipse were kept constant for each tract across subjects. Subsequently, the anatomical accuracy of the ROI placement was validated by a board-certified neuroradiologist. The selected structures, following the nomenclature of Mori et al. (2005), and corresponding ROI size in voxels included the: (i) uncinate fasciculus (UF, 41 voxels), (ii) anterior corona radiata (ACR, 20 voxels), (iii) the anterior aspect of the inferior longitudinal fasciculus (ILF, 27 voxels), (iv) genu of the corpus callosum (55 voxels), (v) the cingulum bundle (CB, 52 voxels) and (vi) the superior longitudinal fasciculus (SLF, 27 voxels). These ROIs were selected based on previous literature associating these structures with memory or attention (Landfield, 1988; Casey et al., 2000; Levin, 2003; Fan et al., 2005; Sperling, 2007). Mean and standard deviation (SD) of FA values for each ROI were recorded.
|
Inter-rater and intra-rater reliability were assessed using the intraclass correlation coefficient (ICC). Two raters measured FA of ROIs in the ACR bilaterally and in the UF bilaterally for 15 normal control subjects and 15 TBI subjects to test inter-rater reliability in the two tracts found to have statistically significant correlation with memory and attention performance. One of these raters repeated these measurements to determine intra-rater reliability. The ICC was first calculated separately for measurements on normal subjects and TBI subjects for both inter- and intra-rater reliability. Subsequently, an overall ICC was determined by pooling the normal and TBI subjects together.
Assessment of attentional control
The Attention Network Task (ANT) (Fan et al., 2005) was used to provide a quantitative assessment of attentional control. The test consists of three cue conditions with two target conditions. The three cue conditions were no-cue (baseline), centre-cue (temporally informative) and a spatial-cue (temporally and spatially informative). Briefly, stimuli consisted of a row of five visually presented arrows each individually pointing either leftward or rightward, against a gray background. The subjects were to indicate the direction of the central arrow by pressing a button with the index finger of the left hand for the left direction or the right hand for the right direction. Either all five arrows pointed in the same direction (the congruent target condition) or the flanking arrows pointed in the opposite direction of the centre arrow (the incongruent condition). Reaction times were recorded and the difference in reaction times between the incongruent and congruent conditions, the conflict score, was used as the measure of attentional control (i.e. congruent reaction time minus incongruent reaction time). Since incongruent trials elicit longer reaction times than congruent trials, the conflict score is expected to be negative-valued. A shorter difference in reaction times of the two trials indicates better attentional control performance and therefore a ‘less negative’ conflict score.
Assessment of memory performance
The California Verbal Learning Test, Second Edition (CVLT-II) was administered to measure memory performance (Delis et al., 2000). An examiner orally presented two word lists (list A and list B) that each contained 16 items, comprised of four words from each of four semantic categories. First there are five presentations of list A, each immediately recalled orally by the examinee. This is followed by a presentation of list B once followed by an immediate recall by the examinee. After a 20-min delay during which non-verbal tasks are administered, the examinee is asked to recall as many words as possible from list A. This last trial, the long delay free recall trial (LDFR), was used as the primary measure of memory performance. A greater number of correctly recalled words indicates better memory performance. The short delay free recall test (SDFR), which entails immediately recalling the list of words, was used to differentiate between encoding and retrieval mechanisms. Raw scores from the CVLT-II were used for each of the memory measures.
Statistics
Non-parametric statistical tests with an alpha of 0.05 were used to determine significance. The Spearman's rho statistic was used to test for significance of correlations. A Bonferroni correction was used to account for multiple comparisons. The alpha corrected for multiple comparisons is reported in the ‘Results’ section. Additionally, partial correlations were used to control for time since injury in the TBI group and age in both the normal control and TBI subjects. These included correlations in normal control subjects between FA (of various structures) and LDFR and FA and conflict score. One normal control subject was removed from any statistical tests pertaining to LDFR because he was a statistical outlier on the memory measure. In the TBI patients, correlations were computed for FA and LDFR, FA and conflict score, lateralization and conflict score, LDFR and SDFR, and finally, FA and SDFR. Subsequently, a partial correlation between UF FA and LDFR was performed while controlling for SDFR. The Mann–Whitney U-test was used to test statistical significance when comparing FA and cognitive performance between the normal control group and mild TBI group.
A multiple linear regression analysis was used to establish a correlational double dissociation, as shown previously for reading skill and digit recall in children (Niogi and McCandliss, 2006). This is done using a series of multiple regression analyses. The aim of the analysis was to assess the unique variance accounted for by attentional control in the left ACR (the dominant hemisphere contributing to attention) after controlling for age and memory performance. This same approach was then applied to the UF to assess the unique variance of memory accounted for by the average FA of the bilateral UF after controlling for age and attentional control. Since both hemispheres of the UF had significant association with memory, the bilateral average was used for this test.
In more detail, each multiple regression analysis consisted of two blocks of independent variables and FA as the dependent measure. The first block contained age and the presumptively irrelevant cognitive measure as independent variables. In the case of the left ACR, the control measures were age and LDFR. In the case of the bilateral UF, the control measures were age and conflict score. The second block contained these control variables with the addition of the pertinent parameter that loads onto the specific region. In the case of the left ACR, the pertinent cognitive measure was conflict score, while, in the bilateral UF, the pertinent parameter was LDFR. Significance of the 
R2 change was computed using an F-statistic.
Post hoc, the previous correlational tests in both the normal and TBI groups were repeated substituting ADC for FA. All statistical tests were performed using SPSS v.14 (SPSS Inc., Chicago, IL, USA).
|
Results |
|---|
|
TopSummaryIntroductionMethodsResultsDiscussionConclusionFundingReferences |
|---|
Structure–function relationships in normal adults I: memory domain
Conventional MRI findings were normal for all control subjects. MRI findings for the TBI patients are summarized in Table 1.
The ICC for inter-rater reliability of ROI measurements in normal subjects was 0.84. For TBI subjects, the inter-rater ICC was 0.84 as well. The ICC for intra-rater reliability of ROI measurements in normal subjects was 0.93. In TBI subjects, the intra-rater ICC was 0.88. When assessing overall reliability across all subjects (normal and TBI), the ICC for inter-rater reliability was 0.84 and for intra-rater reliability was 0.91. As these ICC values were each greater than 0.80, they constitute excellent agreement for both inter- and intra-rater reliability.
The goal of the first analysis was to determine which white matter tract is primarily associated with the memory domain. To do so, we tested whether normal variation of white matter microstructural integrity accounts for inter-individual differences in the Long Delay Free Recall (LDFR) sub-test of the California Verbal Learning Test, Second Edition (CVLT-II) (Delis et al., 2000). The LDFR sub-test measures verbal memory, specifically the ability to recall a list of 16 words after 20 min of non-verbal testing. In this test, the number of correctly recalled words is an index of memory performance. Previous literature suggests that the memory domain is associated with the hippocampus and prefrontal cortex (Landfield, 1988; Levin, 1990; Sperling, 2007). As such, major tracts associated with these areas including the UF, ILF and the genu of the corpus callosum (Fig. 1A–C) were subject to ROI analysis on FA maps from the DTI scan.
We tested whether FA correlates with the LDFR score in normal adults. We found that the average FA of the right and left UF correlated significantly with memory performance (r = 0.61, P = 0.003, Fig. 2A). Additionally, the FA of the right UF significantly correlated with memory performance (r = 0.66, P = 0.001). While the left hemisphere UF did not have a statistically significant correlation, a trend existed (r = 0.42, P = 0.054). The ILF and genu failed to correlate significantly with LDFR in these subjects (left ILF: r = –0.17, P = 0.45; right ILF: r = –0.053, P = 0.81; genu = –0.01, P = 0.97). Table 2 summarizes these results and shows correlations controlling for age. Controlling for multiple comparisons using a Bonferroni corrected alpha level, correlations below P < 0.01 are significant. Using this criterion, a significant correlation exists between the right UF and LDFR.
|
|
Structure–function relationships in normal adults II: attentional control domain
In this analysis, we test which white matter tract is primarily associated with attentional control. Attentional control was measured using the ANT described in detail in the Methods section (Fan et al., 2002
). Briefly, the conflict score is the difference in reaction times between congruent and incongruent flanking arrow trials and serves as an index of attentional control. Conflict scores are expected to be negative to reflect the cost of cognitive conflict, thus greater cognitive conflict is reflected in lower values.
In contrast to the first analysis, we selected ROIs in white matter tracts associated with the anterior cingulate gyrus, which is associated with attentional control (Casey et al., 2000; Fan et al., 2005). We correlated the FA from ROIs in the anterior corona radiata (ACR), the cingulum bundle (CB) and the superior longitudinal fasciculus (SLF) with the conflict scores in normal adults (Fig. 1D–F). Attentional control performance correlated significantly with FA of the left hemisphere ACR (r = 0.42, P = 0.047, Fig. 2B). However, the right hemisphere ACR FA and the bilateral average failed to correlate significantly (right ACR: r = 0.21, P = 0.332; bilateral ACR: r = 0.32, P = 0.138). The other ROIs also did not approach significance (left CB: r = –0.070, P = 0.751; right CB: r = –0.32, P = 0.135; left SLF: r = 0.31, P = 0.149; right SLF: r = 0.061, P = 0.783). Table 3 summarizes these results and shows correlations controlling for age. Controlling for multiple comparisons using a Bonferroni corrected alpha level of P < 0.0083, only nonsigificant trends remain, suggesting a potential relationship between attentional control and the left hemisphere ACR.
|
Structure–function relationships in mild TBI subjects: memory and attentional control
In the next phase of the study we aimed to test whether these results extend to the damaged range. To do so, we first compare the TBI patient group to the normal controls to test whether certain TBI patients perform outside the normal range in memory and attention performance and whether they have FA below the normal range in the UF and ACR. A Mann–Whitney U-test shows that, as a group, the LDFR score and average UF FA are significantly reduced compared with the normal control group (LDFR: U = 278, P = 0.007; UF: U = 276, P = 0.003). While the TBI group was not significantly different than the normal control group in attentional control (conflict score: U = 451, P = 0.56) or bilateral average ACR FA (U = 383, P = 0.11), individual patients within the TBI cohort performed outside the normal range. Eight TBI patients had ANT conflict scores outside the range of the normal controls. Two of these subjects and an additional six TBI patients had FA values outside the normal bilateral average ACR FA range (15 patients when the right and left ACR are considered separately which include all eight TBI patients with conflict scores outside the normal range).
Given that certain patients within the TBI cohort perform below the normal range in the cognitive domains and have FA below the normal range, we test whether the aforementioned significant relationships shown in normal adults extend to the broader range of scores exhibited by adults with mild TBI. Driven by the results of the preceding experiments in normal adults, only the UF and ACR were examined in the mild TBI subjects. Again, memory and attention were assessed using the LDFR sub-test of the CVLT-II and the ANT conflict score, respectively.
The results indicate that the relationships seen in normal adults extend into the damaged range. First, FA of the UF in both hemispheres correlated significantly with memory performance in mild TBI subjects (bilateral average UF: r = 0.52, P < 0.001, Fig. 3A; right UF: r = 0.50, P = 0.001; left UF: r = 0.46, P = 0.002). Table 4 summarizes these results and shows correlations controlling for age and time since injury. Controlling for multiple comparisons using a Bonferroni corrected alpha level of P < 0.025, the left and right and left UF remain significantly correlated with LDFR.
|
|
Secondly, the attentional control and ACR FA relationship also existed in the mild TBI population. The bilateral average ACR FA correlated significantly with the conflict score in mild TBI subjects (r = 0.36, P = 0.022). A closer examination of the contributions of each hemisphere shows that the left ACR FA is the primary contributor to this relationship (r = 0.47, P = 0.001, Fig. 3D). The right hemisphere ACR FA did not correlate with conflict significantly (r = 0.11, P = 0.492). To test whether the effect was left lateralized in the mild TBI subjects, a lateralization score was calculated for the ACR FA, which did not correlate with conflict in mild TBI (r = –0.22, P = 0.160). Table 5 summarizes these results and shows correlations controlling for age and time since injury. Controlling for multiple comparisons using a Bonferroni corrected alpha level of P < 0.025, the left ACR remains significantly correlated to conflict.
|
Specificity of structure–function relationships: a double dissociation
The final goal of the study was to test the specificity of the cognitive domains associated with the ACR and UF by attempting to establish a correlational double dissociation within the normal controls and mild TBI subjects, as shown previously for reading skill and digit recall in children (Niogi and McCandliss, 2006
). This is done using a series of multiple regression analyses. The aim of the analysis was to assess the unique variance accounted for by attentional control in the left ACR after controlling for age and memory performance. This same approach was then applied to the UF to assess the unique variance of memory accounted for by the average FA of the bilateral UF after controlling for age and attentional control. This technique is described in detail in the ‘Statistics’ sub-section of the ‘Methods’ section. Briefly, each multiple regression analysis consisted of two blocks of independent variables and FA as the dependent measure. The first block contained age and the presumptively irrelevant cognitive measure as independent variables. The second block contained these control variables with the addition of the pertinent cognitive measure associated with the specific region.
There was only a significant R2 change in the final blocks after adding the pertinent cognitive measure in each case. After the addition of conflict in the multiple regression scheme for the left ACR, there was a significant R2 change (Normal: R2 = 0.20, P = 0.043; TBI: R2 = 0.20, P = 0.004, Table 6). This implies that, even after controlling for age and memory, attentional control still loaded heavily on the left ACR. Likewise, after the addition of LDFR in the multiple regression scheme for the bilateral UF, there was a significant R2 change (Normal: R2 = 0.23, P = 0.03; TBI: R2 = 0.12, P = 0.018, Table 7). Therefore, in this case, differences in UF integrity are specifically related to memory performance even after controlling for age and differences in attentional control.
|
|
Post hoc analysis: is injury in the UF linked to encoding or retrieval problems in memory?
The aforementioned results suggest that damage to the UF is associated with decreased memory performance in adults with mild TBI. However, the cognitive deficit could either be due to inability to encode the information or difficulties in retrieving the information. To address this, a post hoc analysis was performed on the mild TBI subjects.
Memory was assessed using the LDFR measure of the CVLT-II. To distinguish between encoding and retrieval, one can investigate the ability of subjects to immediately recall the same list of words, measured in the SDFR sub-test of the CVLT-II. If subjects perform well in the SDFR test, but poorly in the LDFR test, then the subjects likely have deficits in memory retrieval because the lists were properly encoded for the immediate recall trials. However, if subjects perform poorly on both the LDFR and SDFR, then the deficit may be in either encoding or both long- and short-term retrieval.
There was no significant correlation of SDFR and LDFR scores in the mild TBI subjects (P = –0.11, P = 0.488). SDFR also does not correlate with FA of the UF (bilateral average UF: r = –0.30, P = 0.851; left UF: r = –0.010, P = 0.953; right UF: –0.108, P = 0.500). Additionally, after controlling for SDFR, a partial correlation between the bilateral FA of the UF and LDFR is significant (0.441, r = 0.006). Furthermore, 75% of mild TBI subjects that scored above 90% in the SDFR test also had LDFR scores below 50%. The data suggest that the deficit in memory performance due to damage in the UF is likely due to difficulties in long-term retrieval, not encoding or short-term retrieval.
Post hoc analysis: structure–function correlations with apparent diffusion coefficient
Structure–function correlations that were statistically significant using FA were repeated using ADC as an exploratory follow-up analysis to better understand the nature of the diffusion changes in the TBI subjects. In normal subjects, no significant correlations existed between ACR ADC and conflict (right ACR: r = –0.32, P = 0.14; left ACR: r = –0.34, P = 0.11; bilateral average: r = –0.38, P = 0.70) nor UF ADC and LDFR (right UF: r = –0.27, P = 0.23; left UF: r = –0.18, P = 0.44; bilateral average: r = –0.25, P = 0.26).
Similarly, in TBI subjects no significant correlations existed between in the structure–function correlations using ADC (right ACR: r = –0.043, P = 0.79; left ACR: r = –0.13, P = 0.44; bilateral average: r = –0.10, P = 0.54) nor UF ADC and LDFR (right UF: r = –0.20, P = 0.22; left UF: r = –0.061, P = 0.70; bilateral average: r = –0.16, P = 0.33).
|
Discussion |
|---|
|
TopSummaryIntroductionMethodsResultsDiscussionConclusionFundingReferences |
|---|
The major finding in this 3T DTI study was that microstructural white matter integrity of bilateral UF is associated with memory performance while integrity of the left ACR is associated with attentional control. Interestingly, these associations account for individual differences of attentional control and memory performance in normal adults. Furthermore, damage to these tracts due to mild TBI is associated with greater deficits in the corresponding cognitive domain. In the final phase of this study, the specificity of these relationships was confirmed by establishing a double dissociation separately in the normal adult group and the TBI patient group. The specificity of these structure–function relationships in both groups suggests DTI can be used as a microstructural imaging biomarker for cognitive function and dysfunction.
Microstructural white matter correlates of memory and attentional control
The presence of such a correlational double dissociation argues strongly for a link between individual differences in specific domains of cognitive performance, such as attention and memory, and individual microstructural differences in specific white matter tract circuitry. This double dissociation excludes potentially confounding factors that might cause spurious correlations between white matter microarchitecture and behavioural measures, such as increased motion artifacts in the DTI scans of subjects who perform poorly on the cognitive tests. The fact that the same relationship was found in normal adults and in TBI also rules out possible sources of bias in comparing the two groups, such as systematic differences between controls and patients in ROI placement within normal-appearing white matter. The dissociation depends on the microstructural integrity of white matter and thus challenges classical notions of purely cortical modules of function. Rather, these findings suggest networks of function comprised of separable sub-systems. In the case of executive function, memory and attentional control are revealed to be two dissociable sub-systems.
Given the role both the frontal lobe and temporal lobe play in memory, it is plausible that the microarchitecture of the UF, which connects the prefrontal cortex and parahippocampal gyrus, would affect memory performance. Other DTI studies have also showed a similar relationship. In patients with schizophrenia, loss of white matter integrity along the UF was associated with verbal memory deficits (Nestor et al., 2004). Additionally, our findings suggest deficits in memory performance attributed to UF microstructural injury in both hemispheres is due to retrieval mechanisms under the context of longer delays. The hemispheric encoding/retrieval asymmetry (HERA) model similarly proposes that both right and left prefrontal cortex are involved in retrieval, while encoding primarily activates the left prefrontal cortex (Kapur et al., 1994).
A prior longitudinal study of mild TBI using the ANT showed that deficits in the executive component of attention, as assessed by the conflict score, persist for at least a month post-injury even when the orienting component had recovered (Halterman et al., 2006). We demonstrate that the integrity of the left ACR, which runs adjacent to and also synapses in the anterior cingulate gyrus, was found to account for specific variation in the executive component of attention in both normal volunteers and in mild TBI patients, while the right ACR only produced a non-significant trend. We do not, however, interpret this as direct evidence for a laterality effect as the degree of laterality between left and right ACR was unrelated to attention measures, and the current study may lack sufficient power to resolve similar trends implicated in the right ACR.
Implications for DTI in the study of TBI
TBI is a growing public health concern which affects nearly 1.5 million Americans each year. Worldwide, approximately 57 million individuals worldwide have been hospitalized by one or more episodes of TBI (Langlois et al., 2006) and this number is only expected to increase as the average age of the population rises and as the number of motor vehicles rises around the world. Despite the high incidence of mild TBI, there exists a disproportionately small number of research studies of mild TBI patients with post-concussive syndrome (PCS) (Bigler, 2004). As such, there are a number of contemporary issues in mild TBI that remain unanswered. Most of these issues fall under a common category of how to evaluate and diagnose mild TBI. Findings on conventional non-invasive neuroimaging techniques such as computed tomography (CT) and MR imaging have not reliably correlated with symptoms and are inconsistent at best for predicting functional outcome following mild TBI (Hammoud and Wasserman, 2002; Inglese et al., 2005). The insensitivity of conventional MR imaging in evaluating DAI underscores the need to use more sensitive techniques such as DTI to measure white matter microstructure.
Accurate diagnosis of TBI is a challenge for conventional neuroimaging because of the microstructural nature of the damage. The primary form of injury is DAI, the mechanism of which is unequal rotational and acceleration forces that induce a shear-strain deformation (Gennarelli and Graham, 1998; Medana and Esiri, 2003; Povlishock and Katz, 2005). These forces cause disruptions of axolemma and neurofilament organization which are thought to impair axonal transport (Blumbergs et al., 1989; Arfanakis et al., 2002; Smith et al., 2003). The resulting DAI is difficult to visualize using MRI or CT (Mittl et al., 1994; van der Naalt et al., 1999). Furthermore, focal lesions associated with DAI that are visualized by conventional MR imaging, even at 3T, are not associated with long-term patient functional outcome or neurocognitive status (Scheid et al., 2003; Hughes et al., 2004; Scheid et al., 2006; Niogi et al., 2008). These studies suggest that conventional MR imaging techniques cannot identify mild TBI patients who will have delayed recovery or persistent post-traumatic symptoms. Recent investigations have shown, however, that DTI is more sensitive to axonal injury than conventional MR imaging techniques in TBI patients (Arfanakis et al., 2002; Huisman et al., 2004; Inglese et al., 2005; Newcombe et al., 2007) and in a mouse model of TBI (Mac Donald et al., 2007).
Perhaps it is not surprising that the two most common cognitive impairments in mild TBI, attentional control and memory, are associated with the ACR and UF, respectively. Both the ACR and UF are very susceptible to rotational shear injury because these rostrally located tracts are subject to the highest rotational forces (Gennarelli and Graham, 1998; Povlishock and Katz, 2005). In a recent study, Niogi et al. found in a cohort of 34 mild TBI patients that the ACR and UF were the two most commonly damaged white matter tracts out of 22 examined throughout the brain (Niogi et al., 2008). The results of the present investigation suggest that the reason memory and executive dysfunction dominate cognitive sequelae after mild TBI is because these cognitive domains are associated with the most frequently damaged white matter structures. A recent neurocognitive study of mild TBI (Malojcic et al., 2008) also demonstrated persistent deficits in both visual attention and short-term memory; however, the authors attributed the results in both cognitive domains to a slowing of general information processing. In contradistinction, our data show that deficits in visual attention and short-term memory are separable and are associated with distinct information processing networks subserved by different white matter pathways. In a previous DTI study, we have shown that the global extent of white matter injury throughout the brain correlates with the speed of general cognitive processing, as measured by simple reaction times to visual stimuli (Niogi et al., 2008). Our current study extends this finding to provide evidence that DTI can be used to differentiate long-term impairments in separable cognitive domains, further validating DTI as a potential biomarker for cognitive dysfunction in TBI.
It was postulated by Fork et al. (2005) that memory deficits should be interpreted as executive dysfunction because there was no indication of deficits in encoding new items. In contrast, our results suggest that rather than implicating a global executive dysfunction syndrome, LDFR memory impairment is more specific and independent of attentional dysfunction. In subjects where there was damage to the left ACR and sparing of the bilateral UF, attentional control impairments tend to predominate. Similarly, in cases where there was relatively more damage to the bilateral UF than the left ACR, memory deficits were more severe.
In a recent DTI study of TBI, Kraus et al. (2007) examined a group of 20 mild TBI patients and demonstrated that these subjects had reduced FA in the cortico-spinal tract, sagittal stratum, and superior longitudinal fasciculus. In their study, a 1 SD threshold below the control mean FA was used to indicate reduced anisotropy. Using this criterion, Kraus et al. (2007) found the number of structures with reduced FA correlated with executive, attentional and memory cognitive measures. While Kraus and colleagues demonstrated that the global extent of damage impacts these cognitive domains, our current study demonstrates that injury to specific networks impairs memory and attentional control domains separately and that normal variation in the microstructure of these tracts also influences cognitive performance. Furthermore, these cognitive deficits exist on a continuous spectrum of severity that is dependent on the degree of microstructural white matter injury, as opposed to categorical loss of function.
Methodological considerations
Numerous studies have established FA as a marker for white matter integrity in TBI, demyelinating diseases, and as a correlate of cognitive function in normal populations (Tuch et al., 2005; Mabbott et al., 2006; Mukherjee and McKinstry, 2006; Newcombe et al., 2007; Grieve et al., 2007; Kraus et al., 2007; Niogi et al., 2008). To limit the number of multiple comparisons, only one DTI measure was used for the a priori analysis. Given the strong association between FA and white matter integrity in previous studies, FA was used as the primary index of white matter microstructural integrity. It is important to note that there are other DTI measures as well that may have been employed. The diffusion ellipsoid can be reduced to radial diffusivity (average of the secondary and tertiary eigenvalues) and axial diffusivity (primary eigenvalue). Additionally, the average of all three eigenvalues is the average diffusion coefficient (ADC) which has been shown to increase in TBI (Goetz et al., 2004; Huisman et al., 2004; Shanmuganathan et al., 2004; Newcombe et al., 2007).
As an exploratory post hoc analysis, significant structure–function correlations were repeated with ADC. Interestingly, no significant correlation existed when using ADC although trends approaching significance were apparent in correlations within the normal control group. Perhaps the current study lacks sufficient power to resolve trends implicated by FA. However, these trends did not exist in the TBI population suggesting that FA may be a more sensitive biomarker of diffuse axonal injury.
There are some limitations to this study worth noting for future investigations. First, premorbid functioning is not known for mild TBI patients, although patients with multiple episodes of TBI or with known neuropsychiatric illnesses, alcohol or drug abuse were excluded. Thus, it is not clear whether their cognitive function and white matter FA reflect the effects of mild TBI or existed prior to the injury. While attention, memory and processing speed are cognitive domains that are particularly vulnerable to the effects of mild TBI, gross deficits of intelligence are seldom present. Unfortunately, in almost all cases of persistent PCS, the patients also suffer from depression which will exacerbate commonly noted symptoms such as headaches, attention problems and memory loss (Arcia and Gualtieri, 1994; Hellawell et al., 1999; Bigler, 2004).
Second, it is important to note that many of our subjects that had damage to the left ACR also had damage to the bilateral UF. These subjects tended to have impairments in both the memory and attentional control domains. Yet, the independence of memory and attentional control impairment was established by the double dissociation and was particularly notable in cases for which there was preferential damage to one structure and relative sparing of the other. It is known that TBI produces heterogeneous loci of damage. It is possible that other structures also may have been injured and are implicated in either memory or executive attention performance.
Third, manual ROIs were used to measure FA in white matter tracts. These are limited in that they can only sample a small portion of the fibre pathway. They also suffer from intra-rater and inter-rater variability in placement of ROIs. However, despite these limitations, manual ROI measurements are widely accepted and remain the most commonly used method for DTI quantitation. Of the alternatives, standard methods for spatial normalization of MR imaging data to perform group comparisons in an unbiased fashion do not work for DTI (Jones et al., 2005). No universally accepted method for automated group analysis of DTI has yet been devised. Three-dimensional DTI tractography can sample a larger portion of the white matter pathway than manual ROIs; however, this technique also requires manual ROI placement to define the tracts. Thus, the method is not unbiased. Also, the ACR and UF are technically difficult to define in their entirety with fibre tracking because both course through regions of crossing pathways for which the tensor model of diffusion is not valid (Hess and Mukherjee, 2007). Moreover, focal haemorrhages or cavities in these tracts due to DAI would also adversely affect fibre tractography (Le et al., 2005). In this study, portions of these tracts were chosen for ROI quantitation that were separate from the regions of major crossing fibres and which appeared normal on conventional 3T MR imaging.
Fourth, our goal was to examine TBI subjects with chronic symptoms and as a result we assessed TBI patients at least 1 month post-injury. The timeline regarding changes in neuropathology (recovery or progression) after TBI is unclear due to a paucity of longitudinal studies. Certain reports suggest that resolution normally occurs within a few days to 3 months post-injury, and a subset continue to have long-term symptoms (Dikmen et al., 1986; Ponsford et al., 2000; Povlishock and Katz, 2005; Kashluba et al., 2008). It is worth noting that the subjects were scanned within a large range of time post-injury (1–53 months). However, correlations did not change significantly after controlling for time elapsed post-injury. Thus, it is likely that those with chronic symptoms have stable lesions that did not resolve. Similarly, the study includes a wide age range (17–61 years). Controlling for age did not significantly alter the correlations, suggesting that chronic memory and attentions deficits following TBI are independent of age-related changes in white matter microstructure.
|
Conclusion |
|---|
|
TopSummaryIntroductionMethodsResultsDiscussionConclusionFundingReferences |
|---|
White matter microstructural integrity of the left ACR is associated with attentional control performance in both normal adults and adults with mild TBI. In addition, the bilateral UF is associated with memory function in both normal adults and adults with mild TBI. Furthermore, results demonstrate domain specificity of these structure–function relationships that reflect a continuous scale of severity rather than a categorical loss of modular function. This study provides evidence that DTI may serve as a microstructural imaging biomarker for cognitive dysfunctions and variations within normal cognitive functions.
|
Funding |
|---|
|
TopSummaryIntroductionMethodsResultsDiscussionConclusionFundingReferences |
|---|
This work was supported by a Collaborative grant from the James S. McDonnell Foundation to the Brain Trauma Foundation with lead investigators from Weill Cornell Medical College, University of California, San Francisco and University of California Berkeley as part of the Cognitive Neurobiological Research Consortium in Traumatic Brain Injury. The first and last authors were supported by NSF grant REC-0337715 (PI:B.D.M.), and a John Merck Scholars Award (PI:B.D.M.).
|
Footnotes |
|---|
*These authors contributed equally to this work.
|
Acknowledgements |
|---|
We would like to thank M. Meeker (UCSF) and R. Sarkar (BTF) for their help in subject recruitment and data collection.
|
References |
|---|
|
TopSummaryIntroductionMethodsResultsDiscussionConclusionFundingReferences |
|---|
Arcia E, Gualtieri CT. Neurobehavioural performance of adults with closed-head injury, adults with attention deficit, and controls. Brain Inj (1994) 8:395–404.[Web of Science][Medline]
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.
Basser PJ, Mattiello J, LeBihan D. MR diffusion tensor spectroscopy and imaging. Biophys J (1994) 66:259–67.[Web of Science][Medline]
Basser PJ, Pierpaoli C. Microstructural and physiological features of tissues elucidated by quantitative-diffusion-tensor MRI. J Magn Reson B (1996) 111:209–19.[CrossRef][Web of Science][Medline]
Bigler ED. Neuropsychological results and neuropathological findings at autopsy in a case of mild traumatic brain injury. J Int Neuropsychol Soc (2004) 10:794–806.[Web of Science][Medline]
Blumbergs PC, Jones NR, North JB. Diffuse axonal injury in head trauma. J Neurol Neurosurg Psychiatry (1989) 52:838–41.
Cabeza R, Nyberg L. Neural bases of learning and memory: functional neuroimaging evidence. Curr Opin Neurol (2000) 13:415–21.[CrossRef][Medline]
Casey BJ, Thomas KM, Welsh TF, Badgaiyan RD, Eccard CH, Jennings JR, et al. Dissociation of response conflict, attentional selection, and expectancy with functional magnetic resonance imaging. Proc Natl Acad Sci USA (2000) 97:8728–33.
Catani M. Diffusion tensor magnetic resonance imaging tractography in cognitive disorders. Curr Opin Neurol (2006) 19:599–606.[Web of Science][Medline]
Delis DC, Kramer JH, Kaplan E, Ober BA. California Verbal Learning Test-Second Edition. (2000) San Antonio, TX: Psychological Corporation.
Dikmen S, McLean A Jr, Temkin NR, Wyler AR. Neuropsychologic outcome at one-month postinjury. Arch Phys Med Rehabil (1986) 67:507–13.[Web of Science][Medline]
Durston S, Casey BJ. What have we learned about cognitive development from neuroimaging? Neuropsychologia (2006) 44:2149–57.[CrossRef][Web of Science][Medline]
Fan J, McCandliss BD, Fossella J, Flombaum JI, Posner MI. The activation of attentional networks. Neuroimage (2005) 26:471–9.[CrossRef][Web of Science][Medline]
Fan J, McCandliss BD, Sommer T, Raz A, Posner MI. Testing the efficiency and independence of attentional networks. J Cogn Neurosci (2002) 14:340–7.[CrossRef][Web of Science][Medline]
Fork M, Bartels C, Ebert AD, Grubich C, Synowitz H, Wallesch CW. Neuropsychological sequelae of diffuse traumatic brain injury. Brain Inj (2005) 19:101–8.[Web of Science][Medline]
Gennarelli TA, Graham DI. Neuropathology of the head injuries. Semin Clin Neuropsychiatry (1998) 3:160–75.[Medline]
Goetz P, Blamire A, Rajagopalan B, Cadoux-Hudson T, Young D, Styles P. Increase in apparent diffusion coefficient in normal appearing white matter following human traumatic brain injury correlates with injury severity. J Neurotrauma (2004) 21:645–54.[CrossRef][Web of Science][Medline]
Grieve SM, Williams LM, Paul RH, Clark CR, Gordon E. Cognitive aging, executive function, and fractional anisotropy: a diffusion tensor MR imaging study. AJNR Am J Neuroradiol (2007) 28:226–35.
Gruber SA, Rogowska J, Holcomb P, Soraci S, Yurgelun-Todd D. Stroop performance in normal control subjects: an fMRI study. Neuroimage (2002) 16:349–60.[CrossRef][Web of Science][Medline]
Halterman CI, Langan J, Drew A, Rodriguez E, Osternig LR, Chou LS, et al. Tracking the recovery of visuospatial attention deficits in mild traumatic brain injury. Brain (2006) 129:747–53.
Hammoud DA, Wasserman BA. Diffuse axonal injuries: pathophysiology and imaging. Neuroimaging Clin N Am (2002) 12:205–16.[CrossRef][Web of Science][Medline]
Hellawell DJ, Taylor RT, Pentland B. Cognitive and psychosocial outcome following moderate or severe traumatic brain injury. Brain Inj (1999) 13:489–504.[CrossRef][Web of Science][Medline]
Hess C, Mukherjee P. Visualizing white matter pathways in the living human brain: Diffusion tensor imaging and beyond. Neuroimaging Clin N Am (2007) 17:407–26.[CrossRef][Web of Science][Medline]
Hughes DG, Jackson A, Mason DL, Berry E, Hollis S, Yates DW. Abnormalities on magnetic resonance imaging seen acutely following mild traumatic brain injury: correlation with neuropsychological tests and delayed recovery. Neuroradiology (2004) 46:550–8.[Web of Science][Medline]
Huisman TA, Schwamm LH, Schaefer PW, Koroshetz WJ, Shetty-Alva N, Ozsunar Y, et al. Diffusion tensor imaging as potential biomarker of white matter injury in diffuse axonal injury. AJNR Am J Neuroradiol (2004) 25:370–6.
Inglese M, Makani S, Johnson G, Cohen BA, Silver JA, Gonen O, et al. Diffuse axonal injury in mild traumatic brain injury: a diffusion tensor imaging study. J Neurosurg (2005) 103:298–303.[Web of Science][Medline]
Jiang H, van Zijl PC, Kim J, Pearlson GD, Mori S. DtiStudio: resource program for diffusion tensor computation and fiber bundle tracking. Comput Methods Programs Biomed (2006) 81:106–16.[CrossRef][Web of Science][Medline]
Jones DK, Symms MR, Cercignani M, Howard RJ. The effect of filter size on VBM analyses of DT-MRI data. Neuroimage (2005) 26:546–54.[CrossRef][Web of Science][Medline]
Kapur S, Craik FI, Tulving E, Wilson AA, Houle S, Brown GM. Neuroanatomical correlates of encoding in episodic memory: levels of processing effect. Proc Natl Acad Sci USA (1994) 91:2008–11.
Kashluba S, Hanks RA, Casey JE, Millis SR. Neuropsychologic and functional outcome after complicated mild traumatic brain injury. Arch Phys Med Rehabil (2008) 89:904–11.[CrossRef][Web of Science][Medline]
Kraus MF, Susmaras T, Caughlin BP, Walker CJ, Sweeney JA, Little DM. White matter integrity and cognition in chronic traumatic brain injury: a diffusion tensor imaging study. Brain (2007) 130:2508–19.
Landfield PW. Hippocampal neurobiological mechanisms of age-related memory dysfunction. Neurobiol Aging (1988) 9:571–9.[Web of Science][Medline]
Langlois JA, Rutland-Brown W, Wald MM. The epidemiology and impact of traumatic brain injury: a brief overview. J Head Trauma Rehabil (2006) 21:375–8.[Web of Science][Medline]
Le TH, Mukherjee P, Henry RG, Berman JI, Ware M, Manley GT. Diffusion tensor imaging with three-dimensional fiber tractography of traumatic axonal shearing injury: an imaging correlate for the posterior callosal ‘disconnection’ syndrome: case report. Neurosurgery (2005) 56:189.[Web of Science][Medline]
Levin HS. Memory deficit after closed head injury. J Clin Exp Neuropsychol (1990) 12:129–53.[Web of Science][Medline]
Levin HS. Neuroplasticity following non-penetrating traumatic brain injury. Brain Inj (2003) 17:665–74.[CrossRef][Web of Science][Medline]
Levine B, Black SE, Cabeza R, Sinden M, McIntosh AR, Toth JP, et al. Episodic memory and the self in a case of isolated retrograde amnesia. Brain (1998) 121(Pt 10):1951–73.
Lundin A, de Boussard C, Edman G, Borg J. Symptoms and disability until 3 months after mild TBI. Brain Inj (2006) 20:799–806.[CrossRef][Web of Science][Medline]
Mabbott DJ, Noseworthy M, Bouffet E, Laughlin S, Rockel C. White matter growth as a mechanism of cognitive development in children. Neuroimage (2006) 33:936–46.[CrossRef][Web of Science][Medline]
Mac Donald CL, Dikranian K, Song SK, Bayly PV, Holtzman DM, Brody DL. Detection of traumatic axonal injury with diffusion tensor imaging in a mouse model of traumatic brain injury. Exp Neurol (2007) 205:116–31.[CrossRef][Web of Science][Medline]
Malojcic B, Mubrin Z, Coric B, Susnic M, Spilich GJ. Consequences of mild traumatic brain injury on information processing assessed with attention and short-term memory tasks. J Neurotrauma (2008) 25:30–7.[CrossRef][Web of Science][Medline]
Medana IM, Esiri MM. Axonal damage: a key predictor of outcome in human CNS diseases. Brain (2003) 126:515–30.
Medina D, Detoledo-Morrell L, Urresta F, Gabrieli JD, Moseley M, Fleischman D, et al. White matter changes in mild cognitive impairment and AD: a diffusion tensor imaging study. Neurobiol Aging (2005) 27:663–72.[CrossRef][Web of Science][Medline]
Milner B, Petrides M, Smith ML. Frontal lobes and the temporal organization of memory. Hum Neurobiol (1985) 4:137–42.[Web of Science][Medline]
Mittl RL, Grossman RI, Hiehle JF, Hurst RW, Kauder DR, Gennarelli TA, et al. Prevalence of MR evidence of diffuse axonal injury in patients with mild head injury and normal head CT findings. AJNR Am J Neuroradiol (1994) 15:1583–9.[Abstract]
Mori S, Wakana S, Van Zijl PCM. MRI atlas of human white matter (2005) Amsterdam, The Netherlands; San Diego, CA: Elsevier.
Moseley M, Bammer R, Illes J. Diffusion-tensor imaging of cognitive performance. Brain Cogn (2002) 50:396–413.[CrossRef][Web of Science][Medline]
Mukherjee P, McKinstry RC. Diffusion tensor imaging and tractography of human brain development. Neuroimaging Clin N Am (2006) 16:19–43. vii.[Web of Science][Medline]
Nestor PG, Kubicki M, Gurrera RJ, Niznikiewicz M, Frumin M, McCarley RW, et al. Neuropsychological correlates of diffusion tensor imaging in schizophrenia. Neuropsychology (2004) 18:629–37.[CrossRef][Web of Science][Medline]
Newcombe VF, Williams GB, Nortje J, Bradley PG, Harding SG, Smielewski P, et al. Analysis of acute traumatic axonal injury using diffusion tensor imaging. Br J Neurosurg (2007) 21:340–8.[CrossRef][Web of Science][Medline]
Niogi SN, McCandliss BD. Left lateralized white matter microstructure accounts for individual differences in reading ability and disability. Neuropsychologia (2006) 44:2178–88.[CrossRef][Web of Science][Medline]
Niogi SN, Mukherjee P, Ghajar J, Johnson C, Kolster RA, Sarkar R, et al. Extent of Microstructural white matter injury in postconcussive syndrome correlates with impaired cognitive reaction time: a 3T diffusion tensor imaging study of mild traumatic brain injury. AJNR Am J Neuroradiol (2008) 29:967–73.
Ponsford J, Willmott C, Rothwell A, Cameron P, Kelly AM, Nelms R, et al. Factors influencing outcome following mild traumatic brain injury in adults. J Int Neuropsychol Soc (2000) 6:568–79.[CrossRef][Web of Science][Medline]
Povlishock JT, Katz DI. Update of neuropathology and neurological recovery after traumatic brain injury. J Head Trauma Rehabil (2005) 20:76–94.[Web of Science][Medline]
Scheid R, Preul C, Gruber O, Wiggins C, von Cramon DY. Diffuse axonal injury associated with chronic traumatic brain injury: evidence from T2*-weighted gradient-echo imaging at 3 T. AJNR Am J Neuroradiol (2003) 24:1049–56.
Scheid R, Walther K, Guthke T, Preul C, von Cramon DY. Cognitive sequelae of diffuse axonal injury. Arch Neurol (2006) 63:418–24.
Schmithorst VJ, Wilke M, Dardzinski BJ, Holland SK. Cognitive functions correlate with white matter architecture in a normal pediatric population: a diffusion tensor MRI study. Hum Brain Mapp (2005) 26:139–47.[CrossRef][Web of Science][Medline]
Shanmuganathan K, Gullapalli RP, Mirvis SE, Roys S, Murthy P. Whole-brain apparent diffusion coefficient in traumatic brain injury: correlation with Glasgow Coma Scale score. AJNR Am J Neuroradiol (2004) 25:539–44.
Shu SY, Wu YM, Bao XM, Leonard B. Interactions among memory-related centers in the brain. J Neurosci Res (2003) 71:609–16.[CrossRef][Web of Science][Medline]
Smith DH, Meaney DF, Shull WH. Diffuse axonal injury in head trauma. J Head Trauma Rehabil (2003) 18:307–16.[Web of Science][Medline]
Sperling R. Functional MRI studies of associative encoding in normal aging, mild cognitive impairment, and Alzheimer's disease. Ann N Y Acad Sci (2007) 1097:146–55.[CrossRef][Web of Science][Medline]
Stuss DT, Alexander MP. Executive functions and the frontal lobes: a conceptual view. Psychol Res (2000) 63:289–98.[CrossRef][Web of Science][Medline]
Stuss DT, Benson DF. Neuropsychological studies of the frontal lobes. Psychol Bull (1984) 95:3–28.[CrossRef][Web of Science][Medline]
Stuss DT, Kaplan EF, Benson DF, Weir WS, Naeser MA, Levine HL. Long-term effects of prefrontal leucotomy–an overview of neuropsychologic residuals. J Clin Neuropsychol (1981) 3:13–32.[Medline]
Tuch DS, Salat DH, Wisco JJ, Zaleta AK, Hevelone ND, Rosas HD. Choice reaction time performance correlates with diffusion anisotropy in white matter pathways supporting visuospatial attention. Proc Natl Acad Sci USA (2005) 102:12212–7.
van der Naalt J, Hew JM, van Zomeren AH, Sluiter WJ, Minderhoud JM. Computed tomography and magnetic resonance imaging in mild to moderate head injury: early and late imaging related to outcome. Ann Neurol (1999) 46:70–8.[CrossRef][Web of Science][Medline]
Vendrell P, Junque C, Pujol J, Jurado MA, Molet J, Grafman J. The role of prefrontal regions in the Stroop task. Neuropsychologia (1995) 33:341–52.[CrossRef][Web of Science][Medline]
CiteULike 
Connotea 
Del.icio.us What's this?
This article has been cited by other articles:
|
|
 |
|
  W. L. Maxwell, M.-A. MacKinnon, J. E. Stewart, and D. I. Graham Stereology of cerebral cortex after traumatic brain injury matched to the Glasgow Outcome Score Brain, November 6, 2009; (2009) awp264v1. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||