Microstructural abnormalities of white matter
differentiate pediatric and adult-onset bipolar
Evidence for white-matter abnormalities in bipolar
disorder (BD) was first observed in deep white-
matter hyperintensities in T2-weighted anatomical
sion of genes involved in the regulation of oligo-
dendrocytes provides further evidence supporting
the possibility of white-matter abnormalities in BD
(3). Diffusion tensor imaging (DTI) is specifically
suited to interrogate the integrity of white matter by
capitalizing on structural components restricting
the diffusion of water (4, 5). Neuronal membrane
and myelin sheaths are cellular structures that
restrict the diffusion of water. In white matter
where the neuronal membrane of axons forms
Lu LH, Zhou XJ, Fitzgerald J, Keedy SK, Reilly JL, Passarotti AM,
Sweeney JA, Pavuluri M. Microstructural abnormalities of white
matter differentiate pediatric and adult onset bipolar disorder.
Bipolar Disord 2012: 14: 597–606. ? 2012 The Authors.
Journal compilation ? 2012 John Wiley & Sons A ⁄S.
Objectives: White-matter microstructure, known to undergo significant
developmental transformation, is abnormal in bipolar disorder (BD).
Available evidence suggests that white-matter deviation may be more
pronounced in pediatric than adult-onset BD. The present study aimed
to examine how white-matter microstructure deviates from a typical
maturational trajectory in BD.
Methods: Fractional anisotropy (FA) was measured in 35 individuals
presenting with first episode BD (type I) and 46 healthy controls (HC)
(aged 9–42) using diffusion tensor imaging (DTI). Patients were
medication free and close to illness onset at the time of the DTI scans.
Tract-based spatial statistics were used to examine the center of white-
matter tracts, and FA was extracted from nine tracts of interest.
Axial, radial, and mean diffusivity were examined in post-hoc analyses.
Results: The left anterior limb of the internal capsule (ALIC) showed
significantly lower FA in pediatric than adult-onset BD. The lower FA in
BD was due primarily to greater radial, rather than decreased axial,
Conclusions: The ALIC connects the frontal lobes with archistriatum,
thalamus, and medial temporal regions, and alteration in these pathways
may contribute to mood dysregulation in BD. Abnormalities in this
pathway appear to be associated with an earlier onset of illness and thus
may reflect a greater susceptibility to illness.
Lisa H Lua,b, Xiaohong Joe Zhouc,d,
Jacklynn Fitzgeralda, Sarah K
Keedye, James L Reillyf, Alessandra
M Passarottia,e, John A Sweeneyg
and Mani Pavuluria,e
aDepartment of Psychiatry, Pediatric Brain
Research and Intervention Center, University of
Illinois at Chicago,bDepartment of Psychology,
Roosevelt University,cDepartments of Radiology,
Neurosurgery, and Bioengineering,dCenter for
Magnetic Resonance Research,eDepartment of
Psychiatry, University of Illinois at Chicago,
fDepartment of Psychiatry, Northwestern University,
Chicago, IL,gDepartments of Psychiatry and
Pediatrics, University of Texas Southwestern,
Dallas, TX, USA
Key words: affect network – anterior limb of
internal capsule – development – diffusion tensor
imaging – limbic system
Received 11 November 2011, revised and
accepted for publication 18 June 2012
Lisa H. Lu, Ph.D.
Department of Psychiatry
Pediatric Brain Research and Intervention Center
Institute for Juvenile Research
University of Illinois at Chicago
1747 West Roosevelt
Chicago, IL 60608
Bipolar Disorders 2012: 14: 597–606
? 2012 John Wiley and Sons A/S
tubular shapes, water can diffuse down the longitu-
dinal axis of axons more easily than in the radial
direction, in which the direction of diffusion is
restricted by the neuronal membrane and myelin
sheaths. Because DTI measures are derived from
the Brownian motion of water, this technique takes
advantage of axonal membrane shape to tap into
the integrity of structures within the white matter.
Fractional anisotropy (FA) reflects the proportion
of diffusivity along the axial axis of fiber tracts
relative to the radial direction (4, 5). Alterations in
FA may be due to a change in diffusivity in either
the axial or radial direction, and thus it can be
fruitful to examine diffusion in these directions
Prior DTI studies of BD have had mixed
findings. Studies that took a region-of-interest
(ROI) approach focused on prefrontal–limbic
circuits theorized to underlie affective dysregula-
tion in BD. In pediatric BD, these studies have
found lower FA in the anterior corona radiata
(ACR) (6) and superior frontal white matter (7).
In adult BD, findings have been more mixed in
both the direction of findings and location of
abnormality, with some reporting lower FA in the
anterior cingulum (8) and frontal white matter (9),
and others reporting higher FA in the genu of the
corpus callosum (10) and frontal white matter (11)
or no abnormality in these regions (12). Examin-
ations of specific tracts within the prefrontal–
limbic circuit in adults have also yielded mixed
findings, with some reporting lower FA in the
anterior thalamic radiation (13), which makes up
a part of the anterior limb of the internal capsule
(14), and others reporting no difference between
adult BD and healthy controls in the subgenual
cingulate or the amygdalo–hippocampal complex
Studies that examined the entire brain using
either voxel-based morphometry (VBM) or tract-
based spatial statistics (TBSS) found white-matter
abnormalities in regions beyond the frontal–limbic
circuit. In pediatric BD, these included lower FA in
the superior longitudinal fasciculus (16), posterior
corona radiata (17), posterior cingulum (16, 17),
corpus callosum (16, 17), fornix (17), and occipital
white matter (18). Findings in adults are again
mixed, with some studies reporting lower FA in the
corpus callosum (19, 20), posterior thalamic radi-
ation (21), and arcuate fasciculus (21), and others
reporting higher FA in inferior parietal (22) and
occipital white matter (22, 23).
In sum, studies of pediatric BD have tended to
find more consistently lower FA than those of
adult BD. The literature on healthy white-matter
development is a helpful context in which to
consider this pattern of findings. Cross sectional
studies of typically developing individuals have
found higher FA with age in the corpus callo-
sum, internal capsule, thalamic radiations, corona
radiata, arcuate fasciculus, and frontal and tem-
poral white matter (24–26). Lebel and colleagues
(26) examined the maturation rate across com-
missural, association, and projection fibers and
reported that callosal fibers and association fibers
reach maturity by late childhood and adoles-
cence, respectively, while projection fibers con-
tinue to mature into early adulthood. To date, no
one has examined how white-matter microstruc-
ture deviates from a typical maturational trajec-
tory in BD, and whether the inconsistencies in
the BD literature may be partially explained by
deviations from a normal neurodevelopmental
Based on the above, we sought to determine
whether early-onset BD is associated with greater
or more consistent white-matter alterations than
BD of adult onset. We measured FA in medica-
tion-free pediatric and adult-onset BD during the
first episode of illness and investigated FA differ-
ences in BD relative to healthy controls. Because
each tract follows a different normal developmen-
tal trajectory, we first obtained the center of tracts
(skeleton) for each participant, then applied tract
masks and extracted mean FA within each tract
skeleton to evaluate group, age, and interaction
effects at the level of the tract. We hypothesized
that white-matter tracts that connect prefrontal
and limbic regions such as the anterior limb of the
internal capsule (ALIC), ACR, and cingulum
would have lower FA in the BD group (6, 27),
and that these differences would be more evident in
the pediatric-onset BD patients.
The BD group consisted of 35 patients (age range
11–42 years). The healthy control (HC) group
consisted of 46 individuals (age range 9–37 years).
The two groups did not differ in age, gender
distribution, handedness, or intelligence quotient
(IQ) (see Table 1). IQ was estimated in adults using
the Wide Range Achievement Test, 3rd Edition,
Reading subtest, and in children using the Wech-
sler Abbreviated Intelligence Scale, Vocabulary
and Matrix Reasoning subtests. Diagnoses were
based on the Schedule for Affective Disorders and
Schizophrenia for School Aged Children–Present
and Lifetime Version (K-SADS-PL) (28) for
the pediatric population, the Structured Clinical
Lu et al.
Interview for DSM-IV (SCID) (29) for the adult
population, and on all available clinical data
reviewed at consensus diagnosis meetings. Patients
with BD were all type I, had experienced their
first episode of mania within the previous month,
and were unmedicated at the time of scanning.
Mania symptoms were assessed with the Young
Mania Rating Scale (30). Depression symptoms
for pediatric participants were assessed using
the Children’s Depression Rating Scale–Revised
(CDRS-R) (31) and for adult participants, the
Hamilton Rating Scale for Depression (HAM-D)
(32). Comorbid diagnoses and history of prior
medication are reported in Table 2. Approxi-
mately half of the BD patients had one or more
comorbid diagnoses and had previously been
primary care physicians before being referred
to our psychiatry clinic. All participants and
parents of minor subjects provided verbal and
written informed consent according to proce-
dures approved by the University of Illinois at
Chicago Institutional Review Board (Chicago, IL,
Participants underwent magnetic resonance imag-
ing (MRI) scans performed on a 3.0 Tesla GE
Signa HDx scanner (General Electric Health Care,
Waukesha, WI, USA) equipped with a 40 mT ⁄m
gradient subsystem using a quadrature head coil.
The MRI protocol included a multi-slice axial DTI
scan using a customized single-shot echo-planar
imaging (EPI) sequence with eddy current correc-
tion capabilities (33). The key data acquisition
parameters for the DTI scan were repetition time
(TR) = 5200 msec, echo time (TE) = 85.5 msec,
field of view (FOV) = 22 cm, slice thickness =
5 mm, slice gap = 1 mm, k-space matrix = 132 ·
132, imaging matrix = 256 · 256, number of
diffusion gradient directions = 27, b = 0 and
750 sec ⁄mm2
(34), NEX = 2, and total scan
time = 4 min 51 sec.
FA measures the standard deviation among diffu-
sivities along the axial and radial axes of fiber tracts
(35). Alterations in FA between patients and con-
trols reflect diffusivity changes in the axial and ⁄or
Table 1. Demographic and psychiatric characterization of participant
(n = 35)
(n = 46)
(under 18 ⁄18+
Male ⁄female, n
t(79) = 0.73, n.s.
Z = 0.76, n.s.
Z = 0.07, n.s.
Z = 0.35, n.s.
t(79) = 1.32, n.s.
17 ⁄1822 ⁄24
(right ⁄left), n
34 ⁄1 44 ⁄2
Values are presented as mean [standard deviation (SD)] unless
CDRS-R = Children’s Depression Rating Scale–Revised; HAM-
D = Hamilton Rating Scale for Depression; IQ, intelligence
quotient; n ⁄a = not available; n.s. = not significant; YMRS =
Young Mania Rating Scale.
aHandedness is from the Annett Behavioral Handedness Index.
bIQ is an estimate of premorbid intellectual functioning in adults
using the Wide Range Achievement Test, 3rd edition, Reading
subtest. In children, the Wechsler Abbreviated Intelligence
Scale, Vocabulary and Matrix Reasoning subtests was used.
cn = 33.
dn = 21.
en = 14.
Table 2. Comorbid diagnoses and prior medication taken by bipolar dis-
order patients (n = 35)
No. of patients with comorbid diagnosis
Cannabis use history
Cocaine use history
Hallucinogen use history
No. of patients with history of exposure to prior
Second-generation antipsychotic agentsa
Non-stimulant for ADHDe
Values are presented as n (%).
ADHD = attention-deficit hyperactivity disorder; GAD = gener-
alized anxiety disorder; ODD = oppositional defiant disorder;
PTSD = posttraumatic stress disorder.
aRisperidone, aripiprazole, quetiapine, and ziprasidone.
bLithium, divalproex, oxcarbazepine, lamotrigine, carbamaze-
pine, and gabapentin.
cSertraline, bupropion, fluoxetine, escitalopram, and venlafaxine.
dMethylphenidate immediate- and long-acting forms, mixed
amphetamine salts, and dexmethylphenidate.
White matter in bipolar disorder
radial directions. Mean diffusivity represents the
average diffusion coefficients of water molecules in
three orthogonal directions of the diffusion eigen-
vectors and complements the information provided
FA, mean, axial (first or principal eigenvalue), and
radial diffusivity (average of the second and third
eigenvalues) maps were also created for exploratory
analyses by fitting a single tensor model to the raw
diffusion data using FSL’s Diffusion Toolbox v2.0
[FDT; Oxford Centre for Functional Magnetic
Resonance Imaging of the Brain (FMRIB)’s Soft-
ware Library, UK; http://www.fmrib.ox.ac.uk/fsl/
fdt/fdt_dtifit.html; (36)], and then brain extracted
using the Brain Extraction Tool (37). We then used
the Tract-Based Spatial Statistics (TBSS) (38) tool
within FSL to shift the center of white-matter tracts
across subjects into spatial correspondence so that
group differences could be evaluated at the center of
tracts. Subjects’ FA data were first aligned into a
common space using FSL’s nonlinear registration
tool FNIRT [FMRIB’s Nonlinear Registration
Tool (39, 40)], which uses a B-spline representation
of the registration warp field (41). Next, the mean
FA image was created and thinned to create a mean
FA skeleton, which represented the center of all
transformation matrix was combined with an affine
transformation to align the data to 1 · 1 · 1 mm
Montreal Neurological Institute (MNI) 152 stan-
dard space, and the resulting transformation was
applied in one step to individual data to minimize
the loss of resolution due to resampling. This
resulting transformation matrix was applied to
mean, axial, and radial diffusivity maps to bring
all diffusion data into correspondence. To exclude
gray matter voxels and minimize false positives due
edge of the white matter, only voxels on the mean
FA map that exceeded an FA value of 0.30 were
included in the analyses.
To evaluate the potential effects of head move-
ment on DTI data, the MCFLIRT tool within FSL
[Motion Correction using FMRIB’s Linear Image
Registration Tool (42)] was used. The 27 different
gradient directions were acquired as a time series,
similar to functional imaging data. We calculated
the transformation necessary to register each indi-
vidual’s gradient volume (i.e., gradient direction)
to the middle volume in the series to assess the
effects of head movement. There was no difference
between groups in head movement, as measured by
either maximum [BD: 0.29 mm, HC: 0.30 mm;
t(79) = 0.46, n.s.] or mean [BD: 0.17 mm, HC:
0.17 mm; t(79) = 0.06, n.s.] displacement.
Based on prior findings with BD patients (6, 27),
we selected nine tracts using ROI masks available
in FSL, which were created from a standard-space
average of diffusion tensor maps from 81 adults by
hand segmentation and included entire tracts
rather than spheres within tracts (43). These nine
tracts included the corpus callosum (divided into
genu, body, and splenium ROIs), internal capsule
(divided into anterior limb, posterior limb, and
retrolenticular ROIs), external capsule, posterior
thalamic radiation, corona radiata (divided into
anterior, superior, and posterior ROIs), superior
longitudinal fasciculus, cingulum, sagittal stratum,
and corticospinal tracts, yielding a total of 27 ROIs
(left and right for each except the corpus callosum).
These masks were applied to each individual’s
volume in standard space to extract the mean value
of tract skeletons within each mask (FA, mean,
axial, and radial diffusivity). These mean values
were analyzed using a general linear model with
group as a fixed factor and age as a random factor.
Bonferroni corrections were applied to analyses
with FA to control for type I error rates. The
presence of comorbid disorders and prior exposure
to psychotropic medication were entered as cova-
riates in post-hoc analyses to determine if the
difference between BD and HC groups remained
after these factors were controlled. Mean, axial,
and radial diffusivity were analyzed post-hoc in
regions where BD and HC groups differed in FA.
Mean FA values within each tract are reported in
Table 3. We examined the deviation in the devel-
opmental trajectory (i.e., change across age span) of
the BD group from the HC group via group · age
interaction. The deviation between BD and HC
varied significantly across age [F(12,42) = 3.88,
p = 0.0005] in the left ALIC after correcting for
multiple comparisons. The reduction in FA was
greater in the younger than in the older BD patients
(Figs. 1 and 2). This finding remained after the
presence of comorbid diagnosis [F(12,41) = 4.05,
p = 0.0003]andprior
[F(12,41) = 4.22, p = 0.0003] were controlled sta-
tistically as covariates. Tracts that showed trend-
significant without, but not with, experiment-wise
type 1 error protection included the bilateral ante-
rior corona radiata, posterior corona radiata, pos-
terior limb of the internal capsule, retrolenticular
Lu et al.
partof the internal capsule, left sagittal stratum, left
external capsule, and right superior corona radiata
(Fig. 1). No tract had a trend-level main effect of
age, and only one tract had a trend-level main effect
of group (right sagittal stratum).
Post-hoc source and age effects
Axial, radial, and mean diffusivity were examined
in the left ALIC to elucidate the source of
diffusivity that led to deviations in FA development
(age effect) between the groups (group effect). The
age effect in the left ALIC was evaluated using a
general linear model with age as a random factor,
consistent with our other analyses. The effect of age
was significant for mean diffusivity [F(25,55) =
2.48, p = 0.003], axial diffusivity [F(25,55) = 2.12,
p = 0.01], and radial diffusivity [F(25,55) = 1.70,
p = 0.05]. Whether these age effects varied by
group was examined by group · age interaction,
with group as a fixed factor in the general linear
model. There was a trend of higher radial diffusivity
in younger BD patients [F(12,42) = 1.92, p =
0.06] (Fig. 2). The pattern of higher mean diffusivity
and axial diffusivity with agewas similar for the two
subject groups [group · age interaction: mean dif-
fusivity F(12,42) = 0.70, p = 0.74;axial diffusivity
F(12,42) = 0.74, p = 0.70]. This suggests that the
difference between BD and HC across age spans in
FA may result more from alterations in radial than
This was the first study to examine white-matter
deviations in first-episode unmedicated BD from
mid-childhood through early adulthood. The
results showed greater deviations in white-matter
microstructure in the left ALIC in BD with
pediatric onset relative to adult onset. The lower
FA in early-onset BD was primarily due to higher
radial diffusivity. We sought effects that were
Table 3. Fractional anisotropy within each white matter tract of interesta
(n = 35)
(n = 46)
Group · age
Values are presented as mean (standard deviation).
ACR = anterior corona radiata; ALIC = anterior limb of internal
capsule; CC = corpus callosum; CST = corticospinal tract;
EC = external capsule; L = left; PCR = posterior corona radiata;
PLIC = posterior limb of internal capsule; PTR = posterior tha-
lamic radiation; R = right; RLIC = retrolenticular part of internal
capsule; SCR = superior corona radiata; SLF = superior longi-
tudinal fasciculus; SS = sagittal stratum.
aThe group · age interaction term’s associated p-values are
reported before correction for multiple comparisons. No tract had
a trend-level age effect, and only one tract had a trend-level
group effect (SS-R, p = 0.04 before correction for multiple com-
parisons). Uncorrected p-values above 0.10 are indicated with
n.s. (not significant). The Bonferroni-corrected p-value required
for statistical significance was p < 0.0018 (see bold value).
Fig. 1. Skeleton of the center of tracts derived from tract-
based spatial statistics (in green). The left anterior limb of the
internal capsule (in red) has a different trajectory of fractional
anisotropy between bipolar disorder and healthy controls
across the age span. In brown are tracts that showed a trend for
a different trajectory between groups across age (shown are the
bilateral anterior and posterior corona radiata, right superior
corona radiata, bilateral posterior limbs of the internal capsule,
bilateral left retrolenticular part of the internal capsule, and the
left external capsule; the left sagittal stratum is not shown).
A=anterior; L=left; P=posterior; R=right.
White matter in bipolar disorder
evident within entire tracts so that interpretation at
the level of specific tracts was appropriate. The
greater pathology of the ALIC in childhood-onset
cases suggested that the pathophysiology of BD in
childhood may be more severe or perhaps differs in
fundamental ways from that in patients with a later
onset of disease.
Affective dysregulation in BD is believed to be
associated with a compromised integration of
prefrontal–limbic circuitry. Functional imaging
studies of healthy individuals and BD patients
support models of top-down regulation of affect,
with prefrontal systems modulating affective arou-
sal in subcortical structures such as the amygdala
and ventral striatum (44, 45). In BD, the ability of
prefrontal structures to modulate subcortical struc-
tures appears to be compromised. For example,
during incidental processing of affective words, BD
patients show less activation in ventrolateral and
dorsolateral prefrontal regions, accompanied by
more activation in the amygdala relative to healthy
controls (46, 47). This affective dysregulation
model emphasizes the connectivity between differ-
ent neural systems supported by the white-matter
pathways needed for communication between
nodes that comprise the relevant neural systems.
The ALIC contains frontopontine fibers and the
anterior thalamic radiation, which contains limbic
(43, 48, 49). Two limbic circuits pass through the
ALIC (49, 50). The Papez circuit consists of projec-
tions from the hippocampus that connect the mam-
cingulate gyrus. The basolateral limbic circuit con-
sists of projections connecting the orbitofrontal
cortex to the dorsomedial nucleus of the thalamus,
amygdala, and temporal pole. Both of these circuits
pass through the ALIC, which is a major conduit
between the prefrontal cortex and the thalamus.
Thus, compromised white-matter microstructure in
this tract may interfere with the processing of
emotional information by the limbic system. The
involvement of the ALIC in mood regulation has
shown that the ALIC is one of the target sites for
deep-brain stimulation for the treatment of intrac-
table depression (51, 52). Abnormality of the ALIC
white-matter integrity has been documented previ-
ously in BD (13).
Of studies that have directly examined the white-
matter integrity of the ALIC in BD, three studies
reported lower FA (13, 53, 54), while another did
not (6). Those that did find compromised ALIC
integrity examined adult, chronic patients, while
the study that did not find this abnormality
examined pediatric patients. This, at first, seems
to contradict our findings, which indicated more
salient FA compromise in pediatric than adult BD.
However, methodological issues may explain this
10 15 20 25
30 3540 45
510 15 20 25
30 35 40 45
5 101520 25
Fig. 2. Fractional anisotropy (FA), axial diffusivity, radial diffusivity, and mean diffusivity for the left anterior limb of the left
internal capsule (ALIC) are plotted on the y-axis, with age on the x-axis. The group · age interaction for FA is due to radial
diffusivity, which also showed a trend effect of group · age interaction, with a greater discrepancy between bipolar disorder (BD) and
healthy controls (HC) in childhood than in adulthood. There was a significant age effect for axial, radial, and mean diffusivity, with
all participants showing increasing diffusivity in the left ALIC.
Lu et al.
difference. In Pavuluri’s study of pediatric BD, a
ROI approach was pursued where FA was exam-
ined in three ROIs consisting of a small number of
pixels (6). Thus, the entire ALIC was not sampled
as in the present study. Another difference is that
our adult BD participants were seen close to the
time of illness onset and had a limited history of
drug treatment, while adult BD patients in the
literature were typically chronically ill and medi-
cated (13, 53, 54). The medication classes used to
treat BD, including lithium and neuroleptic agents,
have been shown to affect gray and white matter
(55–58). Further studies are needed to address the
potential significance of the medication effects on
white-matter integrity in BD and any possible
progression of abnormalities over the course of
illness, to account for the differences found in some
previous studies of chronic patients relative to our
With regard to the differential patterns of ALIC
compromise in early-onset versus adult-onset BD,
several possible interpretations need to be consid-
ered. One is a developmental view in which there
may be a maturational delay in white-matter
integrity in BD, leading to a greater deficit in
younger patients which ameliorates over time.
Longitudinal studies are needed to test this possi-
bility. Alternatively, our data may reflect patho-
physiological differences between early- and adult-
onset BD, with higher levels of alteration in ALIC
pathways in childhood-onset patients that may
confer an increased risk or vulnerability for illness.
Radial diffusivity may indicate a myelination abnormality in
White matter consists of myelin sheaths and the
axons that they cover. Compromise of white-
matter microstructure can arise from three broad
sources: abnormal myelination, axonal membrane
compromise, or decreased coherence of fiber tracts.
We found that FA deviation in the ALIC of BD
participants is characterized by a relatively greater
alteration in radial diffusivity. In animal models,
dysmyelination has been associated with higher
radial diffusivity (59, 60). In human developmental
studies, projection fibers such as those passing
through the ALIC mature late, relative to callosal
and association fibers, and may not reach maturity
until the third decade of life (26). It is not known
whether the abnormally higher radial diffusivity
seen in the ALIC of BD is related to dysmyelina-
tion, delayed myelination, or decreased coherence
of fiber tracts within the ALIC. Another plausible,
although less likely, contributing factor to lower
FA in pediatric BD is reduced axonal integrity.
It has been shown that the primary determinant of
anisotropy is axonal membranes, although myelin
is known to modulate the degree of anisotropy
(61). A decrease in axial diffusivity is often asso-
ciated with axonal injury (62). However, the BD
and HC groups did not differ in axial diffusivity.
This pattern of higher radial diffusivity but no
change in axial diffusivity, together with the glial
abnormalities reported in BD (3, 63–65), point
toward pathogenesis related to myelin. However,
direct links between cellular changes and DTI
alterations in these disorders remain to be estab-
FA in healthy controls
In healthy individuals, most prior studies on the
FA developmental trajectory have shown increas-
ing FA with age (24, 25, 66–69), although decreas-
ing FA with age (70–73) and no discernable change
with age in certain tracts have also been reported
(72,73). We did not find increasing FA in our HC
group for most tracts. The age range studied may
explain this difference. In the largest sample of 5–
29-year-old healthy participants to date, Lebel and
colleagues showed that for many white-matter
tracts in which FA increases with age, the trajec-
tory is curvilinear, with most changes occurring
before the age of ten (26). The age span for our
subjects was 9–42 years. Thus, we did not have a
sufficient sample of subjects in the age range in
which the most dramatic changes are seen.
increasing FA in the ALIC (74) and in the anterior
thalamic radiation [which passes through the ALIC
(50,69)]. Methodological differences may have con-
tributed to why we did not find increasing FA in the
HC group in the present study. Previous longitudi-
nal studies evaluated voxels within fiber skeletons
using TBSS. When significant age effects were
found, they were limited to circumscribed regions
within tracts. Circumscribed regions of change
within white-matter tracts are difficult to interpret,
given the limitations in our current knowledge of
which specific fibers are located in tract regions. In
contrast to these methods, we used TBSS, but we
superimposed tract masks over TBSS results and
extracted FA values across the entire tract. Thus,
the tract of interest rather thanin small components
of them. This method sacrifices fine localization of
regional alterations within tracts but gains the
ability to make more robust inferences at the tract
level. Interestingly, in one of the previous longitu-
dinal studies, Giorgio and colleagues used tractog-
raphy to identify the corticospinal tract, which
White matter in bipolar disorder
passes through the posterior limb of the internal
capsule (74). They also found a lack of age-related
increase when the entire tract was evaluated,
of the internal capsule that showed increasing FA
with age. Together, our data and those of Giorgio
et al. (74) suggest that caution in interpretation is
warranted when age effects are found for localized
regions within tracts, as they may not generalize to
the entire tract, and our current knowledge of fibers
located within specific locations of tracts is limited.
Furthermore, Lebel and Beaulieu (75) reported
longitudinal data on over 100 participants, and
plotted the percentage of participants for whom
FA increased, decreased, or did not change over
time. Although this study did not evaluate the
ALIC, it was interesting that for the tract that
showed the most protracted increase in FA, only
approximately 50% of the participants demon-
strated increasing FA with age; the rest showed
either no change or decreasing FA. Together with
the above-mentioned limitations in interpreting
age-related alteration that are localized to circum-
scribed regions within tracts, the developmental
FA change in the healthy population is not clear-
cut and should be considered with caution.
Certain limitations of the current study should be
noted. First, we utilized a cross-sectional design to
evaluate deviations in white-matter integrity across
the age span in BD. We do not know if the age
effects shown were developmental in nature and
reflected delayed maturation that ameliorates with
age, or if they reflected an altered maturation of
white-matter pathways in those most likely to have
an early onset of illness during childhood. Second,
from a technical perspective, the spatial resolution
in the DTI acquisition used in this study was
limited, which could lead to partial volume effects
that could bias measurements of FA, and mean,
axial, and radial diffusivity measurements. We
minimized this problem by not analyzing voxels
with potentially biased values, restricting our
analysis to voxels whose primary constituents are
white matter, and then further constrained statis-
tical analyses to centers of tracts, to minimize
potential partial volume effects. Higher-resolution
data acquisition techniques would help to over-
come this limitation in future studies. Third, while
white-matter abnormalities were statistically sig-
nificant after correction for multiple comparisons
only in the left ALIC, alterations in multiple other
tracts were significant before correction for multi-
ple comparisons. Thus, it is possible that white-
matter alterations in the ALIC are more consis-
tently pronounced than those in other white-matter
pathways in BD, rather than that the ALIC is a
highly specific site of neuropathology. Future
studies are needed to evaluate the full extent of
age-related white-matter alterations throughout
the brain in BD.
This research program was supported by funds from the
National Instituteof Health
MH62134, MH092702, K23-RR018638) and the National
Alliance for Research on Schizophrenia and Depression. We
thank Minjie Wu for assistance with figure preparation.
JAS has received grant funding from Janssen and has been a
consultant to Pfizer and Takeda. MP has been a speaker for
Bristol-Myers Squibb. LHL, XJZ, JF, SKK, JLR, and AMP
have no conflicts of interest to report.
1. Pavuluri MN, Birmaher B, Naylor MW. Pediatric bipolar
disorder: a review of the past 10 years. J Am Acad Child
Adolesc Psychiatry 2005; 44: 846–871.
2. Botteron KN, Vannier MW, Geller B, Todd RD, Lee BC.
Preliminary study of magnetic resonance imaging charac-
teristics in 8- to 16-year-olds with mania. J Am Acad Child
Adolesc Psychiatry 1995; 34: 742–749.
3. Tkachev D, Mimmack ML, Ryan MM et al. Oligoden-
drocyte dysfunction in schizophrenia and bipolar disorder.
Lancet 2003; 362: 798–805.
4. Basser PJ, Mattiello J, Le Bihan D. Estimation of the
effective self-diffusion tensor form the NMR spin echo.
J Magn Reson, Ser B 1994; 103: 247–254.
5. Basser PJ, Pierpaoli C. Microstructural and physiological
features of tissues elucidated by quantitative-diffusion-
tensor MRI. J Magn Reson, Ser B 1996; 111: 209–219.
6. Pavuluri MN, Yang S, Kamineni K et al. Diffusion tensor
imaging study of white matter fiber tracts in pediatric
bipolar disorder and attention-deficit ⁄hyperactivity disor-
der. Biol Psychiatry 2009; 65: 586–593.
7. Adler CM, Adams J, DelBello MP et al. Evidence of white
matter pathology in bipolar disorder adolescents experi-
encing their first episode of mania: a diffusion tensor
imaging study. Am J Psychiatry 2006; 163: 322–324.
8. WangF,JackowskiM,KalmarJHet al.Abnormalanterior
cingulum integrity in bipolar disorder determined through
9. Adler CM, Holland SK, Schmithorst V et al. Abnormal
frontal white matter tracts in bipolar disorder: a diffusion
tensor imaging study. Bipolar Disord 2004; 6: 197–203.
10. Yurgelun-Todd DA, Silveri MM, Gruber SA, Rohan ML,
Pimentel PJ. White matter abnormalities observed in
bipolar disorder: a diffusion tensor imaging study. Bipolar
Disord 2007; 9: 504–512.
11. Haznedar MM, Roversi F, Pallanti S et al. Fronto-
thalamo-striatal gray and white matter volumes and
anisotropy of their connections in bipolar spectrum
illnesses. Biol Psychiatry 2005; 57: 733–742.
Lu et al.
12. Beyer JL, Taylor WD, MacFall JR et al. Cortical white
matter microstructural abnormalities in bipolar disorder.
Neuropsychopharmacology 2005; 30: 2225–2229.
13. Lin F, Weng S, Xie B, Wu G, Lei H. Abnormal frontal
cortex white matter connections in bipolar disorder: a DTI
tractography study. J Affect Disord 2011; 131: 299–306.
14. Wakana S, Jiang H, Nagae-Poetscher LM, van Zijl PC,
Mori S. Fiber tract-based atlas of human white matter
anatomy. Radiology 2004; 230: 77–87.
15. Houenou J, Wessa M, Douaud G et al. Increased white
matter connectivity in euthymic bipolar patients: diffusion
tensor tractography between the subgenual cingulate and
the amygdalo-hippocampal complex. Mol Psychiatry 2007;
16. Frazier JA, Breeze JL, Papadimitriou G et al. White
matter abnormalities in children with and at risk for
bipolar disorder. Bipolar Disord 2007; 9: 799–809.
17. Barnea-Goraly N, Chang KD, Karchemskiy A, Howe ME,
Reiss AL. Limbic and corpus callosum aberrations in
adolescents with bipolar disorder: a tract-based spatial
statistics analysis. Biol Psychiatry 2009; 66: 238–244.
18. Kafantaris V, Kingsley P, Ardekani B et al. Lower orbital
frontal white matter integrity in adolescents with bipolar I
disorder. J Am Acad Child Adolesc Psychiatry 2009; 48:
19. Benedetti F, Yeh PH, Bellani M et al. Disruption of white
matter integrity in bipolar depression as a possible
structural marker of illness. Biol Psychiatry 2011; 69:
20. Haller S, Xekardaki A, Delaloye C et al. Combined
analysis of grey matter voxel-based morphometry and
white matter tract-based spatial statistics in late-life bipolar
disorder. J Psychiatry Neurosci 2011; 36: 391–401.
21. Chan WY, Yang GL, Chia MY et al. Cortical and
subcortical white matter abnormalities in adults with
remitted first-episode mania revealed by Tract-Based
Spatial Statistics. Bipolar Disord 2010; 12: 383–389.
22. Wessa M, Houenou J, Leboyer M et al. Microstructural
white matter changes in euthymic bipolar patients: a
whole-brain diffusion tensor imaging study. Bipolar Disord
2009; 11: 504–514.
23. Versace A, Almeida JR, Hassel S et al. Elevated left and
reduced right orbitomedial prefrontal fractional anisot-
ropy in adults with bipolar disorder revealed by tract-
based spatial statistics. Arch Gen Psychiatry 2008; 65:
24. Ashtari M, Cervellione KL, Hasan KM et al. White matter
development during late adolescence in healthy males: a
cross-sectional diffusion tensor imaging study. Neuroimage
2007; 35: 501–510.
25. Giorgio A, Watkins KE, Douaud G et al. Changes in
white matter microstructure during adolescence. Neuroim-
age 2008; 39: 52–61.
26. Lebel C, Walker L, Leemans A, Phillips L, Beaulieu C.
Microstructural maturation of the human brain from
childhood to adulthood. Neuroimage 2008; 40: 1044–1055.
27. Lu LH, Zhou XJ, Keedy S, Reilly JL, Sweeney JA. White
matter microstructure in untreated first episode bipolar
disorder with psychosis: comparison with schizophrenia.
Bipolar Disord 2011; 13: 604–613.
28. Kaufman J, Birmaher B, Brent DA, Ryan ND, Rao U.
K-SADS-PL. J Am Acad Child Adolesc Psychiatry 2000;
29. First M, Spitzer RL, Gibbon M, Williams JBW. Structured
Clinical Interview for DSM–IV Axis I Disoders, Patient
Edition (SCID-P). New York: New York State Psychiatric
30. Young RC, Biggs JT, Ziegler VE, Meyer DA. A rating
scale for mania: reliability, validity and sensitivity. Br J
Psychiatry 1978; 133: 429–435.
31. Poznanski EO, Mokros HB. Children’s Depression Rating
Scale, Revised (CDRS-R). Los Angeles: Western Psycho-
logical Services, 1995.
32. Hamilton M. A rating scale for depression. J Neurol
Neurosurg Psychiatry 1960; 23: 56–62.
33. Zhou X, Maier JK, Reynolds HG. Method to Reduce
Eddy Current Effects in Diffusion-Weighted Echo Planar
Imaging. USA: General Electric, 1999.
34. Poonawalla AH, Zhou XJ. Analytical error propagation in
diffusion anisotropy calculations. J Magn Reson Imaging
2004; 19: 489–498.
35. Le Bihan D, Mangin JF, Poupon C et al. Diffusion tensor
imaging: concepts and applications. J Magn Reson Imag-
ing 2001; 13: 534–546.
36. Smith SM, Jenkinson M, Woolrich MW et al. Advances in
functional and structural MR image analysis and imple-
mentation as FSL. Neuroimage 2004; 23: 208–219.
37. Smith SM. Fast robust automated brain extraction. Hum
Brain Mapp 2002; 17: 143–155.
38. 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.
39. Andersson JLR, Jenkinson M, Smith S. Non-linear Opti-
misation. FMRIB technical report TR07JA1. Oxford:
Oxford Centre for Functional MRI of the Brain,
2007. Available from: http://www.fmrib.ox.ac.uk/analysis/
techrep [accessed May 2012].
40. Andersson JLR, Jenkinson M, Smith S. Non-linear Reg-
istration, aka Spatial Normalisation. FMRIB technical
report TR07JA2. Oxford: Oxford Centre for Functional
MRI of the Brain, 2007. Available from: http://www.fmrib.
ox.ac.uk/analysis/techrep [accessed May 2012].
41. Rueckert D, Sonoda LI, Hayes C, Hill DLG, Leach MO,
Hawkes DJ. Non-rigid registration using free-form defor-
mations: application to breast MRI images. IEEE Trans
Med Imaging 1999; 18: 712–721.
42. Jenkinson M, Bannister P, Brady M, Smith S. Improved
optimization for the robust and accurate linear registration
and motion correction of brain images. Neuroimage 2002;
43. Wakana S, Caprihan A, Panzenboeck MM et al. Repro-
ducibility of quantitative tractography methods applied
to cerebral white matter. Neuroimage 2007; 36: 630–644.
44. Pavuluri MN, Passarotti A. Neural bases of emotional
processing in pediatric bipolar disorder. Expert Rev
Neurother 2008; 8: 1381–1387.
45. Strakowski SM, Delbello MP, Adler CM. The functional
neuroanatomy of bipolar disorder: a review of neuroim-
aging findings. Mol Psychiatry 2005; 10: 105–116.
46. Pavuluri MN, O’Connor MM, Harral EM, Sweeney JA.
An fMRI study of the interface between affective and
cognitive neural circuitry in pediatric bipolar disorder.
Psychiatry Res 2008; 162: 244–255.
47. Passarotti AM, Sweeney JA, Pavuluri MN. Differential
engagement of cognitive and affective neural systems in
pediatric bipolar disorder and attention deficit hyperactiv-
ity disorder. J Int Neuropsychol Soc 2010; 16: 106–117.
48. Zarei M, Johansen-Berg H, Jenkinson M, Ciccarelli O,
Thompson AJ, Matthews PM. Two-dimensional popula-
tion map of cortical connections in the human internal
capsule. J Magn Reson Imaging 2007; 25: 48–54.
49. Gutman DA, Holtzheimer PE, Behrens TE, Johansen-Berg
H, Mayberg HS. A tractography analysis of two deep
White matter in bipolar disorder
brain stimulation white matter targets for depression. Biol Download full-text
Psychiatry 2009; 65: 276–282.
50. Zou LQ, Xie JX, Yuan HS, Pei XL, Dong WT, Liu PC.
Diffusion tensor imaging study of the anterior limb of
internal capsules in neuroleptic-naive schizophrenia. Acad
Radiol 2008; 15: 285–289.
51. Malone DA Jr, Dougherty DD, Rezai AR et al. Deep
brain stimulation of the ventral capsule ⁄ventral striatum
for treatment-resistant depression. Biol Psychiatry 2009;
52. Blomstedt P, Sjoberg RL, Hansson M, Bodlund O, Hariz
MI. Deep brain stimulation in the treatment of depression.
Acta Psychiatr Scand 2011; 123: 4–11.
53. McIntosh AM, Maniega SM, Lymer GK et al. White
matter tractography in bipolar disorder and schizophrenia.
Biol Psychiatry 2008; 64: 1088–1092.
54. Sussmann JE, Lymer GK, McKirdy J et al. White mattter
abnormalities in bipolar disorder and schizophrenia
detected using diffusion tensor magnetic resonance imag-
ing. Bipolar Disord 2009; 11: 11–18.
55. Dill J, Wang H, Zhou F, Li S. Inactivation of glycogen
synthase kinase 3 promotes axonal growth and recovery in
the CNS. J Neurosci 2008; 28: 8914–8928.
56. Jones LD, Payne ME, Messer DF et al. Temporal lobe
volume in bipolar disorder: relationship with diagnosis and
antipsychotic medication use. J Affect Disord 2009; 114:
57. Blumberg HP, Krystal JH, Bansal R et al. Age, rapid-
cycling, and pharmacotherapy effects on ventral prefrontal
cortex in bipolar disorder: a cross-sectional study. Biol
Psychiatry 2006; 59: 611–618.
58. Atmaca M, Ozdemir H, Cetinkaya S et al. Cingulate gyrus
volumetry in drug free bipolar patients and patients treated
with valproate or valproate and quetiapine. J Psychiatr Res
2007; 41: 821–827.
59. Song SK, Yoshino J, Le TQ et al. Demyelination increases
radial diffusivity in corpus callosum of mouse brain.
Neuroimage 2005; 26: 132–140.
60. Song SK, Sun SW, Ramsbottom MJ, Chang C, Russell J,
Cross AH. Dysmyelination revealed through MRI as
increased radial (but unchanged axial) diffusion of water.
Neuroimage 2002; 17: 1429–1436.
61. Beaulieu C. The basis of anisotropic water diffusion in the
nervous system – a technical review. NMR Biomed 2002;
62. Song SK, Sun SW, Ju WK, Lin SJ, Cross AH, Neufeld
AH. Diffusion tensor imaging detects and differentiates
axon and myelin degeneration in mouse optic nerve after
retinal ischemia. Neuroimage 2003; 20: 1714–1722.
63. Uranova NA, Vostrikov VM, Orlovskaya DD, Rachma-
nova VI. Oligodendroglial density in the prefrontal cortex
in schizophrenia and mood disorders: a study from the
Stanley Neuropathology Consortium. Schizophr Res 2004;
64. Vostrikov VM, Uranova NA, Orlovskaya DD. Deficit of
perineuronal oligodendrocytes in the prefrontal cortex in
schizophrenia and mood disorders. Schizophr Res 2007;
65. Ongur D, Drevets WC, Price JL. Glial reduction in the
subgenual prefrontal cortex in mood disorders. Proc Natl
Acad Sci USA 1998; 95: 13290–13295.
66. Barnea-Goraly N, Menon V, Eckert M et al. White matter
development during childhood and adolescence: a cross-
sectional diffusion tensor imaging study. Cereb Cortex
2005; 15: 1848–1854.
67. Schmithorst VJ, Wilke M, Dardzinski BJ, Holland SK.
Correlation of white matter diffusivity and anisotropy with
age during childhood and adolescence: a cross-sectional
diffusion-tensor MR imaging study. Radiology 2002; 222:
68. Snook L, Paulson LA, Roy D, Phillips L, Beaulieu C.
Diffusion tensor imaging of neurodevelopment in children
and young adults. Neuroimage 2005; 26: 1164–1173.
69. Bava S, Thayer R, Jacobus J, Ward M, Jernigan TL, Tapert
SF. Longitudinal characterization of white matter matura-
tion during adolescence. Brain Res 2010; 1327: 38–46.
70. Kawaguchi H, Obata T, Ota M et al. Regional heteroge-
neity and age-related change in sub-regions of internal
capsule evaluated by diffusion tensor imaging. Brain Res
2010; 1354: 30–39.
71. Ardekani S, Kumar A, Bartzokis G, Sinha U. Exploratory
voxel-based analysis of diffusion indices and hemispheric
asymmetry in normal aging. Magn Reson Imaging 2007;
72. Bhagat YA, Beaulieu C. Diffusion anisotropy in subcor-
tical white matter and cortical gray matter: changes with
aging and the role of CSF-suppression. J Magn Reson
Imaging 2004; 20: 216–227.
73. Madden DJ, Whiting WL, Huettel SA, White LE, MacFall
JR, Provenzale JM. Diffusion tensor imaging of adult age
differences in cerebral white matter: relation to response
time. Neuroimage 2004; 21: 1174–1181.
74. Giorgio A, Watkins KE, Chadwick M et al. Longitudinal
changes in grey and white matter during adolescence.
Neuroimage 2010; 49: 94–103.
75. Lebel C, Beaulieu C. Longitudinal development of human
brain wiring continues from childhood into adulthood.
J Neurosci 2011; 31: 10937–10947.
Lu et al.