2013;81;25-32 Published Online before print May 24, 2013
Jeremy Strain, Nyaz Didehbani, C. Munro Cullum, et al.
NFL players with concussion history
Depressive symptoms and white matter dysfunction in retired
This information is current as of May 24, 2013
located on the World Wide Web at:
The online version of this article, along with updated information and services, is
Neurology. All rights reserved. Print ISSN: 0028-3878. Online ISSN: 1526-632X.
since 1951, it is now a weekly with 48 issues per year. Copyright © 2013 American Academy of
® is the official journal of the American Academy of Neurology. Published continuously
Jeremy Strain, BS
Nyaz Didehbani, PhD
C. Munro Cullum, PhD
Sethesh Mansinghani, BS
Heather Conover, BS
Michael A. Kraut, MD
John Hart, Jr., MD
Kyle B. Womack, MD
Editorial, page 14
Supplemental data at
Depressive symptoms and white matter
dysfunction in retired NFL players with
Objective: To determine whether correlates of white matter integrity can provide general as well
as specific insight into the chronic effects of head injury coupled with depression symptom
expression in professional football players.
Method: We studied 26 retired National Football League (NFL) athletes who underwent diffusion
tensor imaging (DTI) scanning. Depressive symptom severity was measured using the Beck
Depression Inventory II (BDI-II) including affective, cognitive, and somatic subfactor scores
(Buckley 3-factor model). Fractional anisotropy (FA) maps were processed using tract-based spa-
tial statistics from FSL. Correlations between FA and BDI-II scores were assessed using both
voxel-wise and region of interest (ROI) techniques, with ROIs that corresponded to white matter
tracts. Tracts demonstrating significant correlations were further evaluated using a receiver
operating characteristic curve that utilized the mean FA to distinguish depressed from nonde-
Results: Voxel-wise analysis identified widely distributed voxels that negatively correlated with
total BDI-II and cognitive and somatic subfactors, with voxels correlating with the affective com-
ponent (p , 0.05 corrected) localized to frontal regions. Four tract ROIs negatively correlated
(p , 0.01) with total BDI-II: forceps minor, right frontal aslant tract, right uncinate fasciculus, and
left superior longitudinal fasciculus. FA of the forceps minor differentiated depressed from non-
depressed athletes with 100% sensitivity and 95% specificity.
Conclusion: Depressive symptoms in retired NFL athletes correlate negatively with FA using
either an unbiased voxel-wise or an ROI-based, tract-wise approach. DTI is a promising biomarker
for depression in this population. Neurology?2013;81:25–32
BDI-II 5 Beck Depression Inventory II; CI 5 confidence interval; CTE 5 chronic traumatic encephalopathy; DSM-IV-TR 5
Diagnostic and Statistical Manual of Mental Disorders, 4th edition, text revision; DTI 5 diffusion tensor imaging; FA 5
fractional anisotropy; FLAIR 5 fluid-attenuated inversion recovery; NFL 5 National Football League; ROC 5 receiver oper-
ating characteristic; ROI 5 region of interest; TBI 5 traumatic brain injury; TBSS 5 tract-based spatial statistics.
Depression after traumatic brain injury (TBI) can manifest days or years after injury,1but the
mechanisms underlying this association remain unknown. White matter damage has been
described independently in both major depression and TBI, but whether such damage is etio-
logically associated with mood disturbance in either or both conditions has not been established.
Depressive symptoms can be quantified using self-assessment questionnaires that target defin-
ing characteristics of depression. A popular self-report instrument is the Beck Depression Inven-
tory II (BDI-II), which consists of 21 questions, each rated on a 1–4 scale based on severity.2The
BDI-II provides general information regarding depressive symptoms, but can be further parti-
tioned into subfactors that address different constellations of symptoms. One model proposed by
Buckley et al.3uses the BDI-II to categorize each item into 1 of 3 symptom groupings. The 3
factors are designated as affective (e.g., loss of pleasure, loss of interest), cognitive (e.g., sadness,
self-criticalness), and somatic (e.g., loss of energy, irritability) symptoms.
From the Berman Laboratory for Learning and Memory (J.S., N.D., S.M., H.C., M.A.K., J.H.), Center for Brain Health, Dallas; School of
Behavioral and Brain Sciences (C.M.C., K.B.W.), Department of Psychiatry, The University of Texas at Dallas; Department of Neurology and
Neurotherapeutics (C.M.C., J.H., K.B.W.), University of Texas Southwestern Medical Center, Dallas; and Department of Radiology (M.A.K.),
The Johns Hopkins School of Medicine, Baltimore, MD.
Goto Neurology.org forfull disclosures. Funding information and disclosures deemed relevant by the authors, if any, are provided at the end of the article.
© 2013 American Academy of Neurology25
American-style football players often sus-
tain numerous concussive and subconcussive
impacts—head impacts that do not elicit neu-
rologic symptoms that can lead to white matter
damage.4–6We evaluated a population of
retired National Football League (NFL) players
in order to study the relationship between
white matter integrity and the manifestation
of depressive symptoms. Using diffusion tensor
imaging (DTI), we assessed white matter integ-
rity with the scalar value, fractional anisotropy
(FA),7and correlated that measure with overall
depression scores as well as each of the 3 sub-
factors derived from the BDI-II.
METHODS Subjects. Thirty-two retired NFL athletes under-
went detailed neurologic, neuropsychologic, and neuroimaging
evaluations. Our first subject was enrolled in November 2010
and recruitment is still ongoing. Participants were recruited from
a local gathering of retired NFL athletes living in the north Texas
region, from meetings of the NFL Athletes Association local
chapter, through local advertising, and through word of mouth.
A comprehensive analysis of our entire athlete cohort at an earlier
recruitment stage was performed previously.8Therefore, to alle-
viate complications that could confuse our findings, subjects with
a clinical diagnosis of either mild cognitive impairment or Alz-
heimer disease were excluded, resulting in a sample size of 26.
The athlete cohort ranged in age from 41 to 79 years (mean age
57.8, SD 11.3), with NFL experience ranging from 2 to 15 years
(mean 8.62, SD 3.75). Sixteen were Caucasian and 10 were African
American. Demographic information is shown in table 1, including
former position played classified based on speed, as described in Leh-
man et al.9Concussion history was acquired from self-reports and
classified using the American Academy of Neurology practice param-
eter guidelines for grading concussion (1997). Two subjects reported
having from 1 to 11 concussions (mean 3.85, SD 3.02). On average,
all athletes were scanned 4 weeks following neuropsychological exam-
ination. Anexpandedprofileofoursubjectscanbe foundintablee-1
on the Neurology®Web site at www.neurology.org.
Twenty-two cognitively normal controls were recruited from
prior aging studies. Subjects were excluded if they had prior his-
tory of concussion, repetitive exposure to subconcussional head
injuries, participation in college or professional football, mental
illness, cognitive complaints, or neurologic disorders. Controls
ranged in age from 41 to 77 years (mean 59.4, SD 11.8), with
education ranging from 11 to 20 years (mean 16.2, SD 2.4).
Twenty of the controls were Caucasian, and 2 were African
Standard protocol approvals, registrations, and patient
consents. All subjects gave written informed consent in accor-
dance with the Declaration of Helsinki and the institutional
review boards of the University of Texas Southwestern Medical
Center and University of Texas at Dallas approved the study pro-
tocols and consent forms.
Beck Depression Inventory II. Depression severity was quan-
tified using the BDI-II.2Depressive symptoms were analyzed in 2
formats, with BDI-II total score as well as depressive domains
divided into the 3 subcomponents of affective, somatic, and cog-
nitive symptoms based on the 3-factor model.3Total BDI-II
scores were used to divide the participants into either a nonde-
pressed group with minimal symptoms (1–12) or a depressed
group with mild to moderate symptoms of depression (.12).2
In all cases, a clinical evaluation utilizing the DSM-IV-TR diag-
nostic criteria agreed with the BDI-II–based classification.10
None of the subjects had a history of depression prior to entering
the NFL, and only 2 of the subjects were currently undergoing
treatment for depression at the time of our study. Two trained
neuropsychologists conducted all neuropsychological testing and
were blind to the imaging results at the time of testing.
Statistical analysis of demographic information. Distribu-
tion and median differences between our depressed and nonde-
pressed groups were assessed using a Mann-Whitney U test and
independent-samples median test. Selected variables that were
subjected to these analyses were age, number of concussions,
white matter lesion burden, years in the NFL, and mean FA
for the entire white matter skeleton. All scans were processed
and analyzed by the same individual who was not blind to the
MRI acquisition and analysis. Scanner specifications and
acquisition of our fluid-attenuated inversion recovery (FLAIR)
images and DTI scans are explained in e-Methods along with
our analytical approach to quantifying white matter hyperinten-
sities.8,11Preprocessing of DTI data included correction for
motion and eddy current distortions followed by skull stripping12
using FSL 4.1.7 (www.fmrib.ox.ac.uk/fsl/).13Tensors were
estimated and FA maps created using the MedINRIA software
package (www-sop.inria.fr/asclepios/software/MedINRIA/). The
Table 1Demographic information of our cohort for athletes with (n 5 5) and without (n 5 21) depressiona
Asymptomatic athletes (n 5 21) Symptomatic athletes (n 5 5)
41–79, 58.7 (11.9) 43–62, 54.0 (7.78)
Sex, % male
Years in NFL
2–15, 8.86 (3.98)5–12, 7.6 (2.71)
Beck Depression Inventory II
0–11, 4.29 (3.72) 18–28, 23.6 (4.28)
Number of concussions
0–10, 3.43 (2.87)3–11, 5.6 (3.29)
Abbreviation: NFL 5 National Football League.
aUnless otherwise stated, values are range, mean (SD).
26Neurology 81July 2, 2013
FA data were then processed using tract-based spatial statistics
(TBSS)14in FSL, a technical white matter processing program
that is used for group DTI analysis to a common template. The
TBSS method aligns all subjects to Montreal Neurological
Institute space by creating and applying a nonlinear matrix,
using FSL’s nonlinear registration tool. TBSS creates a study-
specific “group mean FA skeleton” that contains the core
central regions of white matter shared in common by the
subjects. After thresholding the mean FA map at the standard
value 0.2, FA values are projected onto the group skeleton from
each subject for the “local center” of each tract.
correlations of depression severity (as measured by BDI-II total score)
with FA, looking first at the composite BDI-II score and subsequently
at the Buckley 3-factor scores. The voxel-wise correlations were per-
formed using the Randomize tool in FSL utilizing a permutation-
based Monte Carlo analysis with 5,000 permutations, threshold-free
cluster enhancement, and correction for multiple comparisons using
a family-wise error rate of p , 0.05. Age was treated as a covariate
and extracted from each design matrix prior to the analysis. We
assessed for FA differences between nondepressed athletes and cogni-
tively normal controls by performing a voxel-wise group comparison
using age as a covariate and identical parameters as described in the
Tract level analysis. In addition to the voxel-wise analysis, we
examined these data in a tract-wise manner by isolating portions
of the TBSS-derived skeleton that fell within the boundaries of
several regions of interest (ROI) representing specific white mat-
ter tracts. We recruited 9 cognitively normal college students in
order to delineate the best representation of each tract from a nor-
mal population. To construct the white matter tracts we listed in
table 2, we performed a multiple ROI approach in the determin-
istic tractography program called MedINRIA (www-sop.inria.fr/
asclepios/software/MedINRIA/). All white matter tracts were
warped into common space using the same warp matrices
derived from that corresponding subject’s FA map applied with
FLIRT from FSL. In common space all 9 representations of the
same tract were superimposed, and the tract ROIs we defined
reflect the voxels that were present in a majority of the subjects.
All voxels were inspected and edited to ensure that they were
uniquely represented in only one tract.
The mean FA value was calculated for each tract by averaging
skeletal voxels that resided within each tract ROI.15Correlations
were performed between tract-derived FA values and either total
BDI-II score or subfactors of BDI-II. The age-corrected mean FA
values by tract were exported and analyzed with a bivariate Pear-
son correlation using the SPSS statistical program. A more strin-
gent threshold was applied to our data to compensate for the
numerous correlations performed, with a 5 0.01.
For those ROIs that were significantly correlated with the
total BDI-II score, we plotted a receiver operating characteristic
(ROC) curve to test the ability of the mean FA of these tracts
to distinguish the depressed from the nondepressed athletes.
Table 2Association between fractional anisotropy and depression within white matter tractsa
Abbreviations: BDI 5 Beck Depression Inventory; CC_A 5 anterior corpus callosum; CC_P 5 posterior corpus callosum;
Cing 5 cingulum; CS 5 corticospinal tract; DC 5 descending cingulum; FAT 5 frontal aslant tract; FMajor 5 forceps major;
FMinor 5 forceps minor; FOF 5 fronto-occipital fasciculus; ILF 5 inferior longitudinal fasciculus; SLF 5 superior longitu-
dinal fasciculus; UF 5 uncinate fasciculus.
aData are displayed as R values.
bp , 0.01.
Neurology 81July 2, 2013 27
The flow chart (figure 1) demonstrates the details of the design.
Sensitivity, specificity, odds ratio, and positive and negative pre-
dictive values with their respective 95% confidence intervals (CI)
were calculated along with a positive likelihood ratio for the best-
performing cut point identified by the ROC analysis.
Voxel-based morphometry. Volumetric processing was con-
ducted using FSL-VBM within FSL.16We isolated only prefron-
tal gray matter using the ROI available from the Harvard Center
for Morphometric Analysis to quantify a metric of prefrontal
atrophy for each subject in our study. These volumes were then
implemented as a covariate into our voxel-wise and tract-wise
designs in subsequent analyses.
RESULTS Five athletes were identified as depressed
(mean BDI-II 23.6, SD 4.28), while 21 athletes were
not (mean BDI-II 4.29, 3.72). Depressed and non-
depressed athletes did not differ in age, experience
in the NFL, number of concussions, or volume of
T2-weighted white matter hyperintensities on FLAIR
MRI scans. FLAIR scans for depressed subjects are dis-
played in figure e-1. All depressed athletes reported at
least 3 concussions (table 1). Four of the 5 depressed
athletes reported symptom onset to have occurred fol-
lowing retirement from the NFL and one reported
depression symptoms immediately following a concus-
sion that ended his career.
Voxel-wise correlations for all athletes revealed neg-
atively correlated voxels dispersed throughout the
white matter skeleton for FA and BDI-II total score
(p , 0.05 corrected) (figure 2A). Similar distributions
were found for the cognitive and somatic Buckley fac-
tors (figure 2, C and D), but the affective component
was more localized to bilateral frontal and right poste-
rior regions (p , 0.05 corrected) (figure 2B). There
were no significant differences in FA between healthy
controls and nondepressed retired athletes (p . 0.05).
Our results remained essentially unaffected after incor-
porating prefrontal atrophy as an additional covariate
into the voxel-wise analysis.
Table 2 shows all of the tracts that were included
in the analysis along with the corresponding Pearson
coefficients. Four white matter tracts had mean FA
values that were significantly correlated with the total
BDI-II score (p , 0.01). These tracts included the
forceps minor (figure 3A), right uncinate fasciculus
(figure 3B), right frontal aslant tract (figure e-2A),
and left superior longitudinal fasciculus (figure e-2B).
No tracts uniquely correlated with any of the subfac-
tors that did not also correlate with the total BDI-II
score. A second correlation was performed for the
forceps minor after removing the impact of prefrontal
atrophy but the correlation remained significant (p ,
0.01, R 5 20.523).
ROC curve analysis of the 4 tracts with significant
associations between mean FA and total BDI-II score
revealed thatthe meanFA ofthe forcepsminorbest dis-
with an area under the curve of 0.9712 (figure 3C). A
mean FA cut point of 0.3896 misclassified only one
subject and yielded a sensitivity of 100% (95% CI
47.8–100), a specificity of 95.2% (95% CI 76.2–
99.9), an odds ratio of 150.3 (95% CI 5.3–4,229), a
a negative predictive value of 100% (95% CI 83.2–
100), and a positive likelihood ratio of 21.
DISCUSSION Our data from this relatively small
cohort demonstrated a significant association between
white matter integrity, as measured by DTI FA, and
the presence as well as severity of depressive symptoms,
a history of concussive or subconcussive impacts. The
data show that constellations of depressive symptoms
are associatedwith general,as wellaslocal,white matter
disruption, depending on the Buckley 3-factor score
used as a correlate. Using the overall BDI-II score
threshold associated with clinical depression, the FA
Figure 1Standards for the reporting of diagnostic accuracy studies
Flow chart of the study to assess the diagnostic accuracy of forceps minor fractional anisot-
ropy (index test) in detecting depression within professional athletes validated with the Beck
Depression Inventory II as the reference standard.
28Neurology 81 July 2, 2013
value ofone specific tract,theforcepsminor,resultedin
100% sensitivity and 95% specificity for identifying in-
dividuals with depression. The factor common to all of
these individuals that is associated with white matter
dysfunction is their remote history of concussions that
occurred prior to the detection of the subjects’
The BDI-II is commonly used in clinical and
research settings and reliably quantifies depressive
symptoms.17Depression can be manifested by a vari-
ety of symptoms that span from emotional to physical
as described in Buckley’s 3-factor model—affective,
cognitive, and somatic. Our findings showed that
depressive symptom factors in this cohort correlated
differentially with FA: the affective component was
associated with focal white matter regions (bilateral
frontal and right posterior), and the cognitive and
somatic factors were associated with widespread white
disruption has been reported to correspond with a
prior history of depression.18,19Interestingly, voxels
that survived a more stringent threshold of p ,
0.02 for the BDI-II total score and cognitive and
somatic factors were seen in frontal regions, including
the forceps minor, as well as additional posterior re-
gions for the BDI-II total score (figure 2).
DTI measures of FA and other markers of white
matter integrity are commonly used for analyses of
group data, although the strong correlations obtained
in this study were robust to the point that the present
analyses could be applied at the single subject level.
The 5 individuals who reported significant depressive
symptoms, both as a group and individually, could
be differentiated from nondepressed subjects using just
their low FA values. This implies that irrespective of
cause, compromised white matter integrity in the fron-
severity. Additionally, the significant association of
Figure 2Voxel-wise correlation between white matter and depression in National Football League athletes
Each panel shows negative correlations between fractional anisotropy values and Beck Depression Inventory total score (A) and affective (B), somatic (C),
and cognitive (D) subfactors. Red voxels represent significant voxels at p , 0.05 corrected for multiple comparisons and yellow voxels represent voxels that
survive p , 0.02 corrected for multiple comparisons. Axial slices are in radiologic orientation with the results thickened for better visibility using the
Neurology 81July 2, 2013 29
BDI scoreswithFA was independentof prefrontal cor-
tical atrophy, even for the forceps minor, making it
unlikely that secondary axonal degeneration from neu-
ronal loss rather than an intrinsic white matter process
fully accounts for the findings. Frontal projections
(forceps minor in particular) are not only susceptible
to disruption with head injury due to anatomical loca-
tion but this dysfunction has been linked physiologi-
cally to a reduction in hemispheric synchrony.20
White matter dysfunction has been reported in
concussion with diffuse axonal injury detected both
pathologically21and in DTI studies.22Most subjects
in this study have histories of multiple concussions.
Onepreviousinvestigationhas demonstrateda general
association between DTI white matter abnormalities
and depression severity.18However, that study did not
assess the correlations between severity of depression
and abnormalities in individual white matter tracts or
in focused ROI. White matter disruption in the hip-
pocampal and prefrontal regions has been implicated
in treatment-resistant depression23,24and early-onset
depression,25,26particularly in the frontal lobes.24In
TBI patients, connectivity within frontal networks
was found to be more susceptible to damage and re-
sulted in decreased connectivity.27If white matter dis-
ruption is the underlying cause of impairment in TBI,
it is conceivable that the decreased connectivity is sec-
ondary to primary white matter degeneration.
Figure 3 Tract-wise correlations between fractional anisotropy and the Beck Depression Inventory total
Both forceps minor (A) and right uncinate fasciculus (B) survived a statistical threshold of p , 0.01. Subjects with depres-
sion are identified in red and nondepressed athletes are designated in blue. (C) Receiver operating characteristic (ROC)
curve for fractional anisotropy–derived values from the forceps minor as a classifier for depression. The blue line is the
actual classifier data plateauing at 100% sensitivity with 95% specificity. AUC 5 area under the curve; BDI 5 Beck
30 Neurology 81July 2, 2013
Considering that intact frontal circuits may be impor-
tantin mood stability and these circuits’ vulnerability to
head injury, our findings involving the forceps minor
are consistent with the literature as a plausible bio-
marker for depression severity in a population with a
history of concussion.
It is not apparent whether the deficits described in
this article could be a result of chronic traumatic
encephalopathy (CTE), since this diagnosis at present
is a pathologic one.28–30The goal of this study was to
address the implications of white matter impairment
and depression symptom expression during the sub-
jects’ natural lifespan; therefore, any connection to
CTE would be speculative. We can neither confirm
nor deny CTE in any of our patients but this does not
invalidate DTI as a possible tool for early detection of
CTE in the future.
A limitation of our study involves the lack of a
direct temporal association between concussions and
depression. Concussion characterization in these stud-
ies is based on symptom recollections years or even
decades after the episode. This can lead to inaccuracies
in reported number or severity of concussions. Four of
the depressed subjects reported depressive complaints
that did not manifest until after retirement from the
NFL, whereas one subject experienced symptom onset
might elicit these symptoms, or such injury might
interact with other factors to contribute to a delayed
response that can manifest with aging. Both of these
clinical profilescould be explained by whitematterdis-
ruption, and the severity or location of the injury
might influence the timing of symptom onset. Our
datadonot address the issueof subconcussive injury,5,6
which is not possible to accurately quantify retrospec-
tively. Previous studies of white matter disease and
aging in depression focused on elderly adults
with microvascular changes on T2-weighted FLAIR
images.31Sincethere was a presumed etiologyfor those
characteristic lesions, the increased risk of depression ap-
vascular disease. With variable associations between con-
cussion and depression in retired players, factors such as
presence and duration of postconcussive symptoms32or
genetic profile (presence of an APOE e4 allele) may con-
dysfunction. To further determine whether the present
associations imply causation, future longitudinal, pro-
spective studies are indicated.
Our main findings were supported by both our
voxel- and tract-wise analyses, implicating the impor-
tance of the forceps minor in association with depressive
nections between the frontal lobes. Prior studies have
alluded to the importance for cross-communication
between the frontal lobes in relation tothe development
of depressive symptoms.33,34Decreased FA represents
white matter disruption, and subjects with the most
profound depressive symptoms also had the lowest FA
we studied is small, white matter integrity changes seen
in these subjects with depression lead to new insights of
possible markers of behavior disturbances in chronic
effects of head trauma.
Jeremy Strain: primary author of the manuscript, processed, analyzed,
and contributed to interpretation of the data. Nyaz Didehbani: assisted
in the conception and design of the study, acquired neuropsychological
data, interpretation of the data, and provided insightful revisions of the
manuscript. C. Munro Cullum: major role in the conception and design
of the study and offered critical revisions of the manuscript, supervised
and constructed the neuropsychological battery that was administered
to our subjects. Sethesh Mansinghani: responsible for acquisition of the
neuroimaging data and assisted with manuscript revisions, contributed
to the processing of FLAIR scans and quantification of white matter
lesion burden. Heather Conover: primary recruiter for this study, assisted
in acquiring demographic information on the subjects. Michael A. Kraut:
provided expertise during the conception and design portion of the study
and critically reviewed drafts of the manuscript, supervised the initial cre-
ation of the neuroimaging battery for the study. John Hart, Jr.: primary
advisor behind the conception and design of the study, performed clinical
evaluations on subjects who met criteria for depression, provided revi-
sions and assistance in manuscript preparation, contributed to interpreta-
tion of the data, contributed to the administrative duties, and obtained
funding for the study. Kyle B. Womack: primary interpreter of the data
and head of the neuroimaging portion for this study, performed data
analysis and reviewed multiple drafts of the manuscript, supervision of
the project, and assistance in administrative duties.
The authors thank the retired NFL athletes who participated in the study;
David Kennedy at the CMA for training data for FIRST; Christian Hasel-
grove, Centre for Morphometric Analysis, Harvard; Bruce Fischl, Martinos
Center for Biomedical Imaging, MGH; Janis Breeze and Jean Frazier,
Child and Adolescent Neuropsychiatric Research Program, Cambridge
Health Alliance; Larry Seidmanand Jill Goldstein,Department ofPsychiatry
of Harvard Medical School; and Barry Kosofsky, Weill Cornell Medical
Supported by the BrainHealth Institute for Athletes at the Center for
BrainHealth, a research center at the University of Texas at Dallas, and
by a National Institute on Aging grant.
J. Strain, N. Didehbani, C.M. Cullum, S. Mansinghani, H. Conover,
M. Kraut, and J. Hart, Jr. report no disclosures. K. Womack is the primary
investigator for the K23 grant. Go to Neurology.org for full disclosures.
Received December 3, 2012. Accepted in final form February 26, 2013.
1. Kreutzer JS, Seel RT, Gourley E. The prevalence and symp-
tom rates of depression after traumatic brain injury: a com-
prehensive examination. Brain Inj 2001;15:563–576.
2. Beck AT, Steer RA, Ball R, Ranieri W. Comparison of
Beck Depression Inventories-IA and -II in psychiatric out-
patients. J Pers Assess 1996;67:588–597.
3.Buckley TC, Parker JD, Heggie J. A psychometric evalua-
tion of the BDI-II in treatment-seeking substance abusers.
J Subst Abuse Treat 2001;20:197–204.
Neurology 81 July 2, 201331
4. Crisco JJ, Fiore R, Beckwith JG, et al. Frequency and
location of head impact exposures in individual collegiate
football players. J Athl Train 2010;45:549–559.
McAllister TW, Ford JC, Ji S, et al. Maximum principal
strain and strain rate associated with concussion diagnosis
correlates with changes in corpus callosum white matter
indices. Ann Biomed Eng 2012;40;127–140.
Koerte IK, Ertl-Wagner B, Reiser M,Zafonte R, Shenton ME.
White matter integrity in the brains of professional soccer
players without a symptomatic concussion. JAMA 2012;308;
Pierpaoli C, Jezzard P, Basser PJ, Barnett A, Chiro DG.
Diffusion tensor MR imaging of the human brain. Radi-
Hart J, Jr., Kraut MA, Womack KB, et al. Neuroimaging of
cognitive dysfunction and depression in aging retired National
Football League Players. JAMA Neurol 2013;1;1–10.
Lehman EJ, Hein MJ, Baron SL, Gersic CM. Neurode-
generative causes of death among retired National Football
League players. Neurology 2012;79:1970–1974.
American Psychiatric Association. Diagnostic and Statisti-
cal Manual of Mental Disorders, 4th ed. Washington, DC:
American Psychiatric Association; 2000.
Marquez de la Plata C, Ardelean A, Koovakkattu D. Mag-
netic response imaging of diffuse axonal injury: quantitative
assessment of white matter lesion volume. J Neurotrauma
Smith SM. Fast robust automated brain extraction. Hum
Brain Mapp 2002;17:143–155.
Smith SM, Jenkinson M, Woolrich MW, et al. Advances
in functional and structural MR image analysis and im-
plementation as FSL. Neuroimage 2004;23:S208–S219.
Smith SM, Jenkinson M, Johansen-Berg H, et al. Tract-
based spatial statistics: voxelwise analysis of multi-subject
diffusion data. Neuroimage 2006;31:1487–1505.
Huang H, Fan X, Weiner M, et al. Distinctive disruption
patterns of white matter tracts in Alzheimer’s disease with
full diffusion tensor characterization. Neurobiol Aging
Douaud G, Smith S, Jenkinson M, et al. Anatomically
related grey and white matter abnormalities in adolescent-
onset schizophrenia. Brain 2007;130;2375–2386.
Dozois DJA, Dobson KS, Ahnberg JL. A psychometric
evaluation of the Beck Depression Inventory-II. Psychol
Tang CY, Eaves E, Dams-O’Connor K, et al. Diffuse
disconnectivity in TBI: a resting state fMRI and DTI
study. Transl Neurosci 2012;3:9–14.
Yuan Y, Zhang Z, Bai F, et al. White matter integrity of
the whole brain is disrupted in first-episode remitted ger-
iatric depression. Neuroreport 2007;18:1845–1849.
Sponheim SR, McGuire KA, Kang SS, et al. Evidence of
disrupted functional connectivity in the brain after com-
bat-related blast injury. Neuroimage 2011;54:S21–S29.
21.Raghupathi R. Cell death mechanisms following traumatic
brain injury. Brain Pathol 2004;14:215–222.
Wang JY, Bakhadirov K, Devous MD, et al. Diffusion
tensor tractography of traumatic diffuse axonal injury.
Arch Neurol 2008;65:619–626.
Peng H, Zheng H, Li L, et al. High-frequency rTMS
treatment increases white matter FA in the left middle
frontal gyrus in young patients with treatment-resistant
depression. J Affect Disord 2012;136:249–257.
Guo W, Liu F, Chen J, et al. Altered white matter integ-
rity of forebrain in treatment-resistant depression: a diffu-
sion tensor imaging study with tract-based spatial statistics.
Prog Neuropsychopharmacol Biol Psychiatry 2012;38:
Bae JN, MacFall JR, Krishnan RR, Payne ME,
Steffens DC, Taylor WD. Dorsolateral prefrontal
cortex and anterior cingulate cortex white matter alter-
ations in late-life depression. Biol Psychiatry 2006;60:
Taylor WD, MacFall JR, Gerig G, Krishnan RR. Struc-
tural integrity of the uncinate fasciculus in geriatric depres-
sion: relationship with age of onset. Neuropsychiatr Dis
Marquez de la Plata C, Garces J, Shokri EK, et al. Deficits
in functional connectivity of hippocampal and frontal lobe
circuits after traumatic axonal injury. Arch Neurol 2011;
McKee AC, Cantu RC, Nowinski CJ, et al. Chronic trau-
matic encephalopathy in athletes: progressive tauopathy
following repetitive head injury. J Neuropathol Exp Neu-
Gavett BE, Stern RA, McKee AC. Chronic traumatic
encephalopathy: a potential late effect of sport-related con-
cussive and subconcussive head trauma. Clin Sports Med
Gavett BE, Cantu RC, Shenton M, et al. Clinical appraisal
of chronic traumatic encephalopathy: current perspectives
and future directions. Curr Opin Neurol 2011;24:
Dotson VM, Zonderman AB, Kraut MA, Resnick SM.
Temporal relationships between depressive symptoms
and white matter hyperintensities in older men and
women. Int J Geriatr Psychiatry 2012;28:66–74.
Smits M, Houston GC, Dippel DW, et al. Microstructural
brain injury in post-concussion syndrome after minor head
injury. Neuroradiology 2011;53:553–564.
Bajwa S, Bermpohl F, Rigonattis SP, Pascual-Leone A,
Boggio PS, Fregni F. Impaired interhemispheric interac-
tions in patients with major depression. J Nerv Ment Dis
Xu K, Jiang W, Ren L, et al. Impaired interhemispheric
connectivity in medication-naïve patients with major
depressive disorder. J Psychiatry Neurosci 2012;37:
32Neurology 81July 2, 2013
DOI 10.1212/WNL.0b013e318299ccf8 Download full-text
2013;81;25-32 Published Online before print May 24, 2013
Jeremy Strain, Nyaz Didehbani, C. Munro Cullum, et al.
Depressive symptoms and white matter dysfunction in retired NFL players with
This information is current as of May 24, 2013
Updated Information &
including high resolution figures, can be found at:
Supplementary material can be found at:
This article cites 33 articles, 2 of which you can access for free
This article has been cited by 1 HighWire-hosted articles:
This article, along with others on similar topics, appears in the
Permissions & Licensing
or in its entirety can be found online at:
Information about reproducing this article in parts (figures,tables)
Information about ordering reprints can be found online: