Psychiatry Research: Neuroimaging 213 (2013) 186–192

Contents lists available at SciVerse ScienceDirect

Psychiatry Research: Neuroimaging

0925-49http://d

n CorrInstitute

E-m

journal homepage: www.elsevier.com/locate/psychresns

Cavum septum pellucidum in pediatric traumatic brain injury

Timothy Silk a,b,n, Richard Beare a,b, Louise Crossley b, Kirrily Rogers b, Louise Emsell d,Cathy Catroppa b, Miriam Beauchamp c, Vicki Anderson b

a Developmental Imaging, Murdoch Childrens Research Institute, Royal Children's Hospital, Melbourne, Australiab Critical Care and Neurosciences, Murdoch Childrens Research Institute, Royal Children's Hospital, Melbourne, Australiac Department of Psychology, University of Montreal, Canadad Translational MRI, Department of Imaging & Pathology, KU Leuven & Radiology, University Hospitals Leuven, Belgium

a r t i c l e i n f o

Article history:Received 21 January 2013Received in revised form21 February 2013Accepted 10 March 2013

Keywords:Traumatic brain injuryPediatricMagnetic resonance imaging (MRI)Cavum septum pellucidum

27/$ - see front matter & 2013 Elsevier Irelanx.doi.org/10.1016/j.pscychresns.2013.03.001

esponding author at: Developmental Imaging,, Royal Children's Hospital, Melbourne 3052, Auail address: tim.silk@mcri.edu.au (T. Silk).

a b s t r a c t

The cavum septum pellucidum (CSP) is a fluid-filled cavity in the thin midline structure of the septumpellucidum. The CSP has been linked to several neurodevelopmental disorders, but it also occurs as aresult of head injury. The aims were to assess the presence and characterization of the CSP in youth withtraumatic brain injury (TBI), to assess whether injury severity or IQ measures were related to CSP size,and to examine brain morphometry changes associated with the CSP size. Ninety-eight survivors of TBIand 34 control children underwent magnetic resonance imaging (MRI). Numerous methods were used todefine the presence and characterization of the CSP including length, classification of abnormally largeCSP, rating of the CSP, and volume. There was no difference in presence of CSP between TBI patients andcontrols; however, there was larger and more severely graded CSP in the patient group. Size of the CSPcorrelated positively with injury severity, and regions that correlated most significantly with CSP sizewere the right entorhinal cortex and bilateral hippocampus. Characterizing the CSP and related brainchanges may provide important information concerning disturbances seen after a TBI.

& 2013 Elsevier Ireland Ltd. All rights reserved.

1. Introduction

The septum pellucidum is a thin midline structure separatingthe lateral ventricles and is a trusted landmark of brain develop-ment in fetal sonography. In normal fetal development, the two thinleaves of the septum pellucidum begin to cleave and form a fluid-filled cavity, known as the cavum septum pellucidum (CSP), fusingback together around the time of birth. While the presence of a CSPafter birth appears to be a normal variant in the population, recentstudies suggest that an enlarged CSP serves as a marker of disruptedneurodevelopment and has been linked to several neurodevelop-mental and neuropsychiatric disorders, including bipolar disorder(Kim et al., 2007), Tourette's syndrome (Kim and Peterson, 2003),obsessive-compulsive disorder (Chon et al., 2010), antisocial per-sonality disorder and psychopathy (Raine et al., 2010), affectivedisorders (Shioiri et al., 1996), and most notably schizophrenia(Degreef et al., 1992a, 1992b; DeLisi et al., 1993; Fukuzako et al.,1996; Dickey et al., 2007; Flashman et al., 2007; Choi et al., 2008). Inaddition to the developmental presence of the CSP, the cavity mayalso be generated from injury, thought to result from acceleration–

d Ltd. All rights reserved.

Murdoch Childrens Researchstralia. Tel.: þ61 3 9936 6708.

deceleration forces causing a tearing (or fenestration). The CSP hasbeen linked with traumatic brain injuries (TBIs), particularly toprofessional boxers (Corsellis et al., 1973; Cabanis et al., 1986;Bogdanoff and Natter, 1989; Jordan, 2000; McCrory, 2002; Orrisonet al., 2009; Handratta et al., 2010), but also fatal traffic accidents(Pittella and Gusmao, 2005). While the physical fenestration of ahead injury resulting in a CSP clearly has different mechanismsfrom that of a developmental presence, patients with both headinjuries and neurodevelopmental disorders share many symptomsthat are attributed to functions of the limbic system and under-perform on many neuropsychological measures.

Little is known about the functional implications of the CSP, butattempts have been made to link the CSP with regional brainmorphometry as well as differences in cognitive/neuropsychologi-cal abilities. A large CSP has been associated with smaller brainvolumes in regions including the left temporal lobe (Rossi et al.,1994; Nopoulos et al., 1996), hippocampus (Rossi et al., 1994; Dickeyet al., 2007), left parahippocampus (Kasai et al., 2004; Takahashiet al., 2008) and bilateral amygdala (Takahashi et al., 2008). A largeCSP has also been associated with lower scores on Verbal IQ,Performance IQ, and the Full Scale IQ indices of the Wechsler AdultIntelligence Scale (Nopoulos et al., 2000), as well as poorer verballearning and memory (Flashman et al., 2007). However, both thesecognitive and structural findings are derived from cohorts ofpatients with schizophrenia.

T. Silk et al. / Psychiatry Research: Neuroimaging 213 (2013) 186–192 187

TBI in children results in many deficits in the cognitive, behavioraland social domains (Anderson et al., 2005). Examination of the CSP isof particular interest in TBI given its associationwith a broad range ofdevelopmental problems as well as the close relationship to thelimbic system (Bodensteiner and Schaefer, 1990; Schaefer et al., 1994;Bodensteiner et al., 1998). The septum pellucidim is itself part ofthe limbic system. Developmentally, it stems from the commissuralplate, along with the corpus callosum, anterior commissure andthe hippocampal commissure (Rakic and Yakovlev, 1968). It is animportant relay station linking many structures including thehypothalamus, hippocampus, amygdala, habenula, and brainstemreticular formation (Bodensteiner et al., 1998). The developmentalfusing of the two leaves of the septum pellucidum is attributed torapid development of the hippocampus, amygdala, septal nuclei,fornix, corpus callosum and other midline structures (Nopoulos et al.,2000; Kim et al., 2007). Therefore, a persistent CSP appears to disruptgrowth of other limbic structures. Many studies have examinedglobal and regional brain volume loss following head injury, and atvarious time points post-injury (frommonths to years). In spite of theheterogeneity of how TBI presents, the integrity of certain brainregions has consistently been reported to suffer as a result of theinjury, primarily frontal, temporal and limbic regions (Gale et al.,2005; Merkley et al., 2008; Warner et al., 2010). A recent studyexamining regional volume loss within the first 6–14 months post-injury found that the highest rates of atrophy (of between 9.8% and15%) were noted bilaterally in the amygdala, hippocampus, thalamus,and putamen (Warner et al., 2010). Given the relatively large volumeatrophy effect in primarily limbic regions, the aims of the currentstudy were to assess the presence and characterization of the CSP inyouth with TBI, to assess whether injury severity or IQ measureswere related to CSP size, and to examine any other brain morpho-metry changes associated with the CSP size. Characterizing the CSPand any related brain changes may provide important informationconcerning anatomic and functional disturbances seen after a TBI.

2. Methods

2.1. Participants

This study comprised 132 children, 98 survivors of TBI (68males) and 34 typically developing (TD) children (21 males),matched for age, sex and socio-economic status. All participantswere ascertained between 2007 and 2010, and were between 5.5and 15 years of age at the time of recruitment. Children with TBIwere recruited at the time of injury, and represented consecutiveadmissions to the Royal Children's Hospital, Melbourne, Australia.Healthy control children were recruited from the community,through local schools chosen to provide a range of socio-economic backgrounds. Children were recruited to participate ina longitudinal study, which aimed to investigate the psychosocialconsequences of TBI.

TBI patients (mean age 10.5572.46 years) underwent a structuralMRI scan on average 38.5716.34 days post-injury. Inclusion criteriawere (i) age at injury 5.5–14 years; (ii) documented evidence of a TBI(only closed head injuries were included), as defined below; (iii)evidence of a period of altered consciousness; and (iv) Englishspeaking. Exclusion criteria were (i) documented history of previoushead injury; (ii) non-accidental injury; and (iii) evidence of a pre-existing physical, neurological, psychiatric, developmental or atten-tional disorder. Patients were categorized according to injury sever-ity, using a combination of injury-related information:

(a)

mild TBI (n¼54): Glasgow Coma Score (GCS) (Teasdale andJennett, 1974) on admission 13–15, indicating some alterationof level of consciousness (e.g., drowsiness, disorientation),

with no evidence of mass lesion on clinical CT/MRI and noneurologic deficits;

(b)

mild/complex TBI (n¼13): GCS 13–15, but evidence of pathol-ogy on CT/MRI.

(c)

moderate TBI (n¼21): GCS on admission 9–12, indicatingsignificantly altered consciousness, with reduced responsive-ness and/or mass lesion or other evidence of specific injury onclinical CT/MRI, and/or neurological impairment; and

(d)

severe TBI (n¼10): GCS on admission 3–8, representing coma,and mass lesion or other evidence of specific injury on clinicalCT/MRI and/or neurological impairment.

Thirty-four healthy control participants were matched in age(mean 10.4172.75 years) to the TBI group.

At 6 months post-injury, each participant completed the two-subtest version of the Wechsler Abbreviated Intelligence Scale(WASI) (Wechsler, 1999), and the FSIQ estimate was calculated(M¼100, S.D.¼15).

2.2. Data acquisition

Neuroimaging data were acquired on the 3 T Siemens TIM Trioscanner (Siemens, Erlangen, Germany) at the Murdoch ChildrensResearch Institute, Royal Children's Hospital, Melbourne. Participantslay supine with their head supported in a 12-channel head coil. Highresolution T1-weighted structural MR images were acquired for eachparticipant (TR¼1900 ms, FA¼91, FOV¼256 � 224, 176 slices, slicethickness¼1.0 mm, in-plane resolution 0.5 � 0.5 mm).

2.3. CSP measures

Numerous methods exist to define the presence and character-ization of the CSP. In the current study, characterization of the CSPused a combination of previously described methods in the literatureincluding (1) CSP length, (2) classification of abnormally large CSP,and (3) rating of the CSP. We also included volume of the CSP. To ourknowledge volumetric measures of the CSP have not been usedbefore, probably due to the small volume of the structure and theproblems with low resolution. Structural MR images were acquiredat a 0.5 � 0.5 mm resolution, allowing volume to be acquired.

2.3.1. CSP lengthExamining MRI slices in the coronal view, the anterior–poster-

ior length of the CSP was measured by the number of contiguousslices containing evidence of the CSP (Nopoulos et al., 1998; Kimand Peterson, 2003; Flashman et al., 2007; Chon et al., 2010).

2.3.2. Abnormally large CSPCSP was classified as ‘abnormally large’ if the CSP length was

equal to or greater than 6 mm (observed on 12 or more slices)(Nopoulos et al., 1997; Kwon et al., 1998; Dickey et al., 2007; Choiet al., 2008; Chon et al., 2010).

2.3.3. CSP ratingScans were rated using a previously published ordinal scale

(Degreef et al., 1992b; Fukuzako et al., 1996; Kim and Peterson,2003; Flashman et al., 2007; Choi et al., 2008; Chon et al., 2010).Ratings were graded on the coronal slice that showed the greatestevidence of a cavity. The rating grades assigned were 0 (absent),1 (questionable), 2 (mild), 3 (moderate), 4 (severe) (see Fig. 1). Fora rating of 0 or 1, it was considered that there was no CSP present.For ratings of 2–4, there was a clear presence of CSP with varyingdegrees of severity. The same observer (TS) rated all scans blind todiagnosis. For 32 random cases, a second observer (LE), also blindto diagnosis, rated CSPs and inter-rater reliability was calculated to

T. Silk et al. / Psychiatry Research: Neuroimaging 213 (2013) 186–192188

be 0.88. Inter-rater disagreement was resolved through discussionand consensus.

2.3.4. CSP volumeCSP volumes were traced manually using itk-SNAP software

(www.itksnap.org). The volume of the CSP is presented as apercentage proportional to the participant’s own intra-cranialvolume (ICV: grey matter, white matter and cerebral spinal fluid).

2.4. Brain morphometry measures

Morphometric analysis was conducted using the FreeSurferimage analysis suite (http://surfer.nmr.mgh.harvard.edu) which

Table 1Cavum septum pellucidum data for patients with traumatic brain injury and healthy co

TBI (N¼98) C

Grade N % N

0 (absent) 10 10.2 21 (questionable) 19 19.4 72 (mild) 44 44.9 23 (moderate) 23 23.5 24 (severe) 2 2 0

Presence of CSP 69 70.4 2

Of those present N ¼ 69 N

Abnormally large CSP 10 14.5 0Grade2 44 63.8 23 23 33.3 24 2 2.9 0

Mean S.D. M

No. of slices (length) 9.5 11.6 6Volume (% of ICV) 1.6 1.6 1

TBI, traumatic brain injury; CSP, cavum septum pellucidum; ICV, intra-cranial volume.n po0.05.

Fig. 1. Examples of the grading system for the cavum septum pellucidum. MRIcoronal slices that showed the greatest evidence of a cavity, with assigned ratinggrades: 0 (absent), 1 (questionable), 2 (mild), 3 (moderate), 4 (severe).

automatically segments and parcellates cortical grey matter andsub-cortical structures. This approach has previously been used inboth adult (Warner et al., 2010) and pediatric (Merkley et al., 2008)TBI. Full details regarding the analysis steps are described else-where (Dale et al., 1999; Fischl et al., 2002, 2004a, 2004b; Jovicichet al., 2009). Using a probabilistic labeling algorithm that relies ongyral and sulcal information, the cortex was parcellated intoanatomical regions of interest (ROIs) based on the Deskman–Killiany atlas and projected back into each subject's native space.From this automated parcellation procedure, we obtained volumemeasurements from a total of 35 cortical and eight sub-corticalgrey matter regions in each hemisphere.

2.5. Statistical analysis

Chi-square tests were used to assess differences between groupsin the presence and grade of abnormally large CSPs. Differencesbetween slice number and CSP volume were assessed using abetween-group t-test. To examine the relationship of CSP volumewith injury severity and other measures of brain morphometry,Pearson correlations were performed. Statistical analyses were two-tailed, and po0.05 was used as the level of significance. For brainmorphometry correlation, a threshold of po0.001 corresponds to athreshold of po0.05 false discovery rate (FDR) corrected for thenumber of multiple comparisons.

3. Results

3.1. CSP presence and categorization

Data on the presence and characteristics of the CSP in the TBI andcontrol groups are presented in Table 1. TBI patients and healthycontrols did not differ in the prevalence of the CSP; however, the TBIgroup exhibited a difference in grading distribution towards greaterseverity of CSP, as measured by the CSP rating. Of those participantsin which a CSP was present, the TBI group exhibited significantlymore abnormally large CSPs than controls, as well as larger volume ofthe CSP. Between-group differences were not significant when CSPlength was compared.

ntrols.

ontrols (N¼34) Analysis

% χ2 p

5.920.6

3 67.65.90

5 73.5 0.13 0.73

¼ 25

0 4.06 0.04*

3 9220 7.21 0.03*

ean S.D. F p

.4 2.4 2.39 0.13.1 0.8 5.58 0.02*

T. Silk et al. / Psychiatry Research: Neuroimaging 213 (2013) 186–192 189

3.2. CSP size and injury severity

To examine the relationship between CSP and injury severityfor the participants with a TBI, a Pearson correlation was per-formed using the CSP volume and the categorical class of injuryseverity. CSP volume showed a significant positive correlationwithinjury severity (r¼ 0.34, p¼0.001) (see Fig. 2).

3.3. CSP size and cognitive measures

Pearson correlational analysis was performed on the CSPvolume in order to detect any correlation with scores on intellec-tual function. There was no significant correlation between FSIQ

Fig. 2. Graph of the cavum septum pellucidum volume correlation with injuryseverity categorization in traumatic brain injury.

Table 2Brain regions whose volume correlates with CSP volume.

Brain region Left

r

Rostral middle frontalPars triangularis −0.158Pars orbitalis −0.158Lateral orbitofrontal −0.247Medial orbitofrontalTemporal pole −0.239Entorhinal cortex −0.189Bank STS −0.207Superior temporalThalamus proper −0.166Caudate −0.224Putamen −0.211Pallidum −0.214Hippocampus −0.299ParahippocampalAmygdala −0.199Accumbens −0.205Ventral diencephalon −0.187Inferior parietal

Note: Correlations reported are significant at po0.05. Trends towards significance are in Significant at false discovery rate of 0.05.

and CSP volume in either the TBI or control group individually orcombined. There were no between-group differences in FSIQ.

3.4. CSP size and other brain structures

To examine whether a relationship could be identified betweenthe CSP volume and other brain morphometrics within the TBIgroup, Pearson correlations were performed between CSP volumeand each of the 86 regions generated from the Freesurfer segmen-tation and parcellation.

Table 2 shows the brain regions correlating with CSP volumestating the direction of the correlation. Nomenclature of brainregions is derived from the Freesurfer parcellation. Trends towardssignificance were reported if po0.08 where the contralateralhomologue was also found to be significant. Only the rightentorhinal cortex and bilateral hippocampus survived FDR-correction for multiple comparisons. Fig. 3 depicts the grey matterregions affected.

4. Discussion

The current study aimed to characterize the CSP seen inpatients after TBI and to assess any related brain changes.Surprisingly, the TBI patients and healthy controls did not differin the prevalence of the CSP, as measured by the ordinal ratingscale defined previously (Degreef et al., 1992b; Fukuzako et al.,1996; Kim and Peterson, 2003; Flashman et al., 2007; Choi et al.,2008; Chon et al., 2010). The prevalence estimates for the CSP evenin normal adults have varied widely, from 2% to 81.6% (e.g. Degreefet al., 1992a; Takahashi et al., 2007), which in part may depend onthe method used to define it. In addition to the grading systemadopted to determine overall size, the methods used includedefining CSP as being present if it is identified on at least oneMRI slice (ranging from 0.45 mm to 1.5 mm thick), the number ofslices on which the CSP appears, and considering a CSP abnormallylarge if it is greater than or equal to 6 mm in length. There havealso been changes in MRI technology over time, such as improved

Right

p r p

−0.176 0.0460.073 −0.197 0.0250.074 −0.163 0.0650.005

−0.217 0.0140.006 −0.204 0.020.031 −0.323 o0.001*0.018

−0.195 0.0270.06 −0.155 0.080.011 −0.222 0.0120.0170.015 −0.155 0.080.001* −0.277 0.001*

−0.174 0.0490.023 −0.181 0.040.02 −0.228 0.0090.034 −0.212 0.016

−0.224 0.011

talicized. Bank STS = bank of the superior temporal sulcus.

Fig. 3. Rendered brain showing grey matter and sub-cortical regions correlating with CSP volume.

T. Silk et al. / Psychiatry Research: Neuroimaging 213 (2013) 186–192190

resolution, which may also contribute to some of the discrepanciesbetween studies.

Despite a lack of findings regarding presence of a CSP using thegrading system, there was a significant difference in the gradingdistribution towards greater severity of CSP in the TBI group.In addition the TBI group exhibited significantly more abnormallylarge CSPs than controls and a greater volume of the CSP. The onlymeasure that did not reveal a between-group difference was thenumber of slices, but this technique only characterizes the CSP inone dimension, along the anterior–posterior direction, and there-fore may not fully characterize the extent of the fenestration.

The presence of the CSP has been previously described as resultingfrom injury, where the acceleration–deceleration forces cause afenestration generating the cavum (Corsellis et al., 1973; Cabaniset al., 1986; Bogdanoff and Natter, 1989; Jordan, 2000; McCrory,2002; Pittella and Gusmao, 2005; Orrison et al., 2009; Handrattaet al., 2010). This has mainly been described in boxers, who suffer froma long-term sustained cause of brain injury; however, the currentstudy examined single incident, non-fatal TBIs, which are often theresult of a fall or car accident, allowing the examination of therelationship between CSP and the severity of the injury. Indeed, CSPvolume showed a significant positive correlation with injury severity,with more severe head injuries generally having larger CSP volumes. Alarger CSP has also previously been associated with lower scores onVerbal, Performance, and the Full Scale IQ indices in schizophrenia(Nopoulos et al., 2000). While TBI patients typically display lowerscores on such tests, there was no between-group difference in FSIQand no significant correlation with CSP volume.

This study also set out to examine the brain volume changesassociated with the CSP volume. While scanning occurred relativelyacutely (average 38 days post-injury), the CSP volume was shown tosignificantly correlate with many cortical and sub-cortical regions,but constrained almost exclusively to inferior prefrontal and tem-poral regions as well as the cerebellum and nearly all sub-corticalstructures. Whether these structures are independently susceptibleto injury or whether there is some causal relationship betweenregions remains to be seen, but many of the structures are closelylinked in their development and function. All regions showed anegative correlation with CSP volume.

These affected regions are concordant with both previous volu-metric studies of TBI (e.g. Gale et al., 2005; Wilde et al., 2005b) andvolumetric correlations with CSP in schizophrenia (Rossi et al., 1994;Nopoulos et al., 1996; Kasai et al., 2004; Dickey et al., 2007;Takahashi et al., 2008). Neuroimaging studies that have documentedbrain atrophy following pediatric TBI have found overall brainatrophy (Bigler, 1999; Wilde et al., 2005a), but also specificallyenlarged ventricle volumes (Verger et al., 2001) and grey and whitematter loss in prefrontal, orbitofrontal, temporal and limbic brainregions (Gale et al., 2005; Tasker et al., 2005; Wilde et al., 2005a;Merkley et al., 2008; Warner et al., 2010). Studies focusing onparticular ROIs have shown marked reductions of hippocampal,amygdala, basal ganglia, and cerebellar volumes (Tasker et al., 2005;

Spanos et al., 2007; Wilde et al., 2007; Beauchamp et al., 2011),suggesting particular vulnerability to injury in children.

The regions that correlated most significantly with CSP volume,and survived FDR correction, were bilateral hippocampus and theright entorhinal cortex. Hippocampal volume reduction is consis-tent with previous findings (Rossi et al., 1994; Dickey et al., 2007),and is also concordant with the finding that the hippocampus is aregionwith one of the highest rates of atrophy post-injury (Warneret al., 2010). The hippocampus has developmentally close ties withthe septum pellucidum, stemming from the same commissuralplate (Rakic and Yakovlev, 1968), and a particular developmentalstep of the septum pellucidum is attributed to rapid developmentof the hippocampus and other midline structures (Nopoulos et al.,2000; Kim et al., 2007). The entorhinal cortex, lying adjacent to thehippocampus, provides the largest source of input to the hippo-campus, and both structures are primary ROIs regarding the onsetof Alzheimer's disease (Li et al., 2012). Given the role in memorythat the entorhinal cortices and hippocampus play, and previousfindings linking CSP with poorer learning and memory (Flashmanet al., 2007), speculatively, the CSP size in a scan following a TBImay be useful as a prognostic indicator of the functional outcomesin the memory domain. Therefore, future studies examining therelationship between CSP volume and the structure and function ofbrain regions thought to be linked to the septum pellucidumwarrant further consideration.

There are some limitations in this study to be considered. First,there were only a small number of patients in the severe headinjury group. Second, injury severity was defined primarily by GCSscore (a measure of consciousness), which may be related to butnot necessarily indicative of the actual force and injury to thebrain. Inherent to research in TBI, no information was availableconcerning the presence and size of the CSP prior to the injury,and given the lack of consensus of the prevalence of CSP in thehealthy population, we cannot completely exclude the possibilitythat the CSP was not present developmentally and perhaps leavesthese brain regions more susceptible to injury or leadsto one being more likely to have an accident (e.g., by increasedrisk-taking behavior or impaired coordination).

In conclusion, the current study failed to find a significantdifference in presence of CSP between TBI patients and healthycontrols; however, there was evidence of larger and more severelygraded CSP in the patient group. The size of the CSP correlatedpositively with injury severity, and brain regions that correlatedmost significantly with CSP size were the right entorhinal cortexand bilateral hippocampus. While the function of the septumpellucidum remains elusive, it forms important links in the limbicsystem and therefore may have great clinical importance.

Acknowledgments

This work was supported by the Victorian Neurotrauma Initia-tive, by the Victorian Government's Operational Infrastructure

T. Silk et al. / Psychiatry Research: Neuroimaging 213 (2013) 186–192 191

Support Program and by the Royal Children's Hospital staff andpatients. TS was supported by an NHMRC Career DevelopmentAward.

References

Anderson, V., Catroppa, C., Morse, S., Haritou, F., Rosenfeld, J., 2005. Attentional andprocessing skills following traumatic brain injury in early childhood. BrainInjury 19, 699–710.

Beauchamp, M.H., Ditchfield, M., Maller, J.J., Catroppa, C., Godfrey, C., Rosenfeld, J.V.,Kean, M.J., Anderson, V.A., 2011. Hippocampus, amygdala and global brainchanges 10 years after childhood traumatic brain injury. International Journalof Developmental Neuroscience 29, 137–143.

Bigler, E.D., 1999. Neuroimaging in pediatric traumatic head injury: diagnosticconsiderations and relationships to neurobehavioral outcome. Journal of HeadTrauma Rehabilitation 14, 406–423.

Bodensteiner, J.B., Schaefer, G.B., 1990. Wide cavum septum pellucidum: A markerof disturbed brain development. Pediatric Neurology 6, 391–394.

Bodensteiner, J.B., Schaefer, G.B., Craft, J.M., 1998. Cavum septi pellucidi and cavumvergae in normal and developmentally delayed populations. Journal of ChildNeurology 13, 120–121.

Bogdanoff, B., Natter, H.M., 1989. Incidence of cavum septum pellucidum in adults:a sign of boxer's encephalopathy. Neurology 39, 991–992.

Cabanis, E.A., Perez, G., Tamraz, J.C., Iba-Zizen, M.T., Roger, B., Alfonso, J.M.,Rougemont, D., 1986. Cephalic magnetic resonance imaging of boxers. Pre-liminary results. Acta Radiologica – Supplementum 369, 365–366.

Choi, J.S., Kang, D.H., Park, J.Y., Jung, W.H., Choi, C.H., Chon, M.W., Jung, M.H., Lee, J.M.,Kwon, J.S., 2008. Cavum septum pellucidum in subjects at ultra-high risk forpsychosis: compared with first-degree relatives of patients with schizophreniaand healthy volunteers. Progress in Neuro-Psychopharmacology & BiologicalPsychiatry 32, 1326–1330.

Chon, M.W., Choi, J.S., Kang, D.H., Jung, M.H., Kwon, J.S., Chon, M.-W., Choi, J.-S.,Kang, D.-H., Jung, M.H., Kwon, J.S., 2010. MRI study of the cavum septumpellucidum in obsessive-compulsive disorder. European Archives of Psychiatry& Clinical Neuroscience 260, 337–343.

Corsellis, J.A., Bruton, C.J., Freeman-Browne, D., 1973. The aftermath of boxing.Psychological Medicine 3, 270–303.

Dale, A.M., Fischl, B., Sereno, M.I., 1999. Cortical surface-based analysis. I. Segmen-tation and surface reconstruction. Neuroimage 9, 179–194.

Degreef, G., Bogerts, B., Falkai, P., Greve, B., Lantos, G., Ashtari, M., Lieberman, J.,1992a. Increased prevalence of the cavum septum pellucidum in magneticresonance scans and post-mortem brains of schizophrenic patients. PsychiatryResearch: Neuroimaging 45, 1–13.

Degreef, G., Lantos, G., Bogerts, B., Ashtari, M., Lieberman, J., 1992b. Abnormalitiesof the septum pellucidum on MR scans in first-episode schizophrenic patients.AJNR: American Journal of Neuroradiology 13, 835–840.

DeLisi, L., Hoff, A., Kushner, M., Degreef, G., 1993. Increased prevalence of cavumseptum pellucidum in schizophrenia. Psychiatry Research: Neuroimaging 45,1–13.

Dickey, C.C., McCarley, R.W., Xu, M.L., Seidman, L.J., Voglmaier, M.M., Niznikiewicz, M.A.,Connor, E., Shenton, M.E., 2007. MRI abnormalities of the hippocampus and cavumsepti pellucidi in females with schizotypal personality disorder. SchizophreniaResearch 89, 49–58.

Fischl, B., Salat, D.H., Busa, E., Albert, M., Dieterich, M., Haselgrove, C., van der Kouwe, A.,Killiany, R., Kennedy, D., Klaveness, S., Montillo, A., Makris, N., Rosen, B., Dale, A.M.,Fischl, B., Salat, D.H., Busa, E., Albert, M., Dieterich, M., Haselgrove, C., van der Kouwe,A., Killiany, R., Kennedy, D., Klaveness, S., Montillo, A., Makris, N., Rosen, B., Dale, A.M., 2002. Whole brain segmentation: automated labeling of neuroanatomicalstructures in the human brain. Neuron 33, 341–355.

Fischl, B., Salat, D.H., van der Kouwe, A.J., Makris, N., Segonne, F., Quinn, B.T., Dale, A.M., Fischl, B., Salat, D.H., van der Kouwe, A.J.W., Makris, N., Segonne, F., Quinn, B.T., Dale, A.M., 2004a. Sequence-independent segmentation of magnetic reso-nance images. Neuroimage 23 (Suppl 1), S69–S84.

Fischl, B., van der Kouwe, A., Destrieux, C., Halgren, E., Segonne, F., Salat, D.H., Busa, E.,Seidman, L.J., Goldstein, J., Kennedy, D., Caviness, V., Makris, N., Rosen, B., Dale, A.M., Fischl, B., van der Kouwe, A., Destrieux, C., Halgren, E., Segonne, F., Salat, D.H.,Busa, E., Seidman, L.J., Goldstein, J., Kennedy, D., Caviness, V., Makris, N., Rosen, B.,Dale, A.M., 2004b. Automatically parcellating the human cerebral cortex. CerebralCortex 14, 11–22.

Flashman, L.A., Roth, R.M., Pixley, H.S., Cleavinger, H.B., McAllister, T.W., Vidaver, R.,Saykin, A.J., 2007. Cavum septum pellucidum in schizophrenia: clinical andneuropsychological correlates. Psychiatry Research: Neuroimaging 154, 147–155.

Fukuzako, T., Fukuzako, H., Kodama, S., Hashiguchi, T., Takigawa, M., 1996. Cavumseptum pellucidum in schizophrenia: a magnetic resonance imaging study.Psychiatry & Clinical Neurosciences 50, 125–128.

Gale, S.D., Baxter, L., Roundy, N., Johnson, S.C., 2005. Traumatic brain injury andgrey matter concentration: a preliminary voxel based morphometry study.Journal of Neurology, Neurosurgery & Psychiatry 76, 984–988.

Handratta, V., Hsu, E., Vento, J., Yang, C., Tanev, K., 2010. Neuroimaging findings andbrain-behavioral correlates in a former boxer with chronic traumatic braininjury. Neurocase 16, 125–134.

Jordan, B., 2000. Chronic traumatic brain injury associated with boxing. Seminars inNeurology 20, 179–185.

Jovicich, J., Czanner, S., Han, X., Salat, D., van der Kouwe, A., Quinn, B., Pacheco, J., Albert,M., Killiany, R., Blacker, D., Maguire, P., Rosas, D., Makris, N., Gollub, R., Dale, A.,Dickerson, B.C., Fischl, B., Jovicich, J., Czanner, S., Han, X., Salat, D., van der Kouwe, A.,Quinn, B., Pacheco, J., Albert, M., Killiany, R., Blacker, D., Maguire, P., Rosas, D.,Makris, N., Gollub, R., Dale, A., Dickerson, B.C., Fischl, B., 2009. MRI-derivedmeasurements of human subcortical, ventricular and intracranial brain volumes:Reliability effects of scan sessions, acquisition sequences, data analyses, scannerupgrade, scanner vendors and field strengths. Neuroimage 46, 177–192.

Kasai, K., McCarley, R.W., Salisbury, D.F., Onitsuka, T., Demeo, S., Yurgelun-Todd, D.,Kikinis, R., Jolesz, F.A., Shenton, M.E., 2004. Cavum septi pellucidi in first-episode schizophrenia and first-episode affective psychosis: an MRI study.Schizophrenia Research 71, 65–76.

Kim, K.J., Peterson, B.S., 2003. Cavum septi pellucidi in Tourette syndrome.Biological Psychiatry 54, 76–85.

Kim, M.J., Lyoo, I.K., Dager, S.R., Friedman, S.D., Chey, J., Hwang, J., Lee, Y.-J., Dunner,D.L., Renshaw, P.F., 2007. The occurrence of cavum septi pellucidi enlargementis increased in bipolar disorder patients. Bipolar Disorders 9, 274–280.

Kwon, J.S., Shenton, M.E., Hirayasu, Y., Salisbury, D., Fischer, I., Dickey, C.C.,Yurgelun-Todd, D., Tohen, M., Kikinis, R., Jolesz, F.A., McCarley, R.W., 1998.MRI study of cavum septi pellucidi in schizophrenia, affective disorder, andschizotypal personality disorder. American Journal of Psychiatry 155, 509–515.

Li, X., Jiao, J., Shimizu, S., Jibiki, I., Watanabe, K., Kubota, T., 2012. Correlationsbetween atrophy of the entorhinal cortex and cognitive function in patientswith Alzheimer's disease and mild cognitive impairment. Psychiatry & ClinicalNeurosciences 66, 587–593.

McCrory, P., 2002. Cavum septi pellucidi—a reason to ban boxers? British Journal ofSports Medicine 36, 157–161.

Merkley, T.L., Bigler, E.D., Wilde, E.A., McCauley, S.R., Hunter, J.V., Levin, H.S.,Merkley, T.L., Bigler, E.D., Wilde, E.A., McCauley, S.R., Hunter, J.V., Levin, H.S.,2008. Diffuse changes in cortical thickness in pediatric moderate-to-severetraumatic brain injury. Journal of Neurotrauma 25, 1343–1345.

Nopoulos, P., Krie, A., Andreasen, N.C., 2000. Enlarged cavum septi pellucidi inpatients with schizophrenia: clinical and cognitive correlates. Journal ofNeuropsychiatry and Clinical Neurosciences 12, 344–349.

Nopoulos, P., Swayze, V., Andreasen, N.C., 1996. Pattern of brain morphology inpatients with schizophrenia and large cavum septi pellucidi. Journal ofNeuropsychiatry and Clinical Neurosciences 8, 147–152.

Nopoulos, P.C., Giedd, J.N., Andreasen, N.C., Rapoport, J.L., 1998. Frequency andseverity of enlarged cavum septi pellucidi in childhood-onset schizophrenia.American Journal of Psychiatry 155, 1074–1079.

Nopoulos, P.C., Swayze, V., Flaum, M., Ehrhardt, J.C., Yuh, W.T., Andreasen, N.C.,1997. Cavum septi pellucidi in normals and patients with schizophrenia asdetected by magnetic resonance imaging. Biological Psychiatry 41, 1102–1108.

Orrison, W.W., Hanson, E.H., Alamo, T., Watson, D., Sharma, M., Perkins, T.G., Tandy,R.D., 2009. Traumatic brain injury: a review and high-field MRI findings in 100unarmed combatants using a literature-based checklist approach. Journal ofNeurotrauma 26, 689–701.

Pittella, J.E., Gusmao, S., 2005. Cleft cavum of the septum pellucidum in victims offatal road traffic accidents: a distinct type of cavum associated with severediffuse axonal injury. Surgical Neurology 63 (S30-34), S34–S35. (discussion).

Raine, A., Lee, L., Yang, Y., Colletti, P., 2010. Neurodevelopmental marker for limbicmaldevelopment in antisocial personality disorder and psychopathy. TheBritish Journal of Psychiatry: The Journal of Mental Science 197, 186–192.

Rakic, P., Yakovlev, P.I., 1968. Development of the corpus callosum and cavum septiin man. Journal of Comparative Neurology 132, 45–72.

Rossi, A., Stratta, P., Mancini, F., Gallucci, M., Mattei, P., Core, L., Di Michele, V.,Casacchia, M., 1994. Magnetic resonance imaging findings of amygdala-anteriorhippocampus shrinkage in male patients with schizophrenia. PsychiatryResearch: Neuroimaging 52, 43–53.

Schaefer, G.B., Bodensteiner, J.B., Thompson Jr., J.N., 1994. Subtle anomalies of theseptum pellucidum and neurodevelopmental deficits. Developmental Medicine& Child Neurology 36, 554–559.

Shioiri, T., Oshitani, Y., Kato, T., Murashita, J., Hamakawa, H., Inubushi, T., Nagata, T.,Takahashi, S., 1996. Prevalence of cavum septum pellucidum detected by MRI inpatients with bipolar disorder, major depression and schizophrenia. Psycholo-gical Medicine 26, 431–434.

Spanos, G.K., Wilde, E.A., Bigler, E.D., Cleavinger, H.B., Fearing, M.A., Levin, H.S., Li,X., Hunter, J.V., 2007. cerebellar atrophy after moderate-to-severe pediatrictraumatic brain injury. American Journal of Neuroradiology 28, 537–542.

Takahashi, T., Suzuki, M., Hagino, H., Niu, L., Zhou, S., Nakamura, K., Tanino, R.,Kawasaki, Y., 2007. Prevalence of large cavum septi pellucidi and its relation tothe medial temporal structures in schizophrenia spectrum. Progress in Neuro-Psychopharmacology & Biological Psychiatry 31, 1235–1241.

Takahashi, T., Yung, A.R., Yücel, M., Wood, S.J., Phillips, L.J., Harding, I.H., Soulsby, B.,McGorry, P.D., Suzuki, M., Velakoulis, D., Pantelis, C., 2008. Prevalence of largecavum septi pellucidi in ultra high-risk individuals and patients with psychoticdisorders. Schizophrenia Research 105, 236–244.

Tasker, R.C., Salmond, C.H., Westland, A.G., Pena, A., Gillard, J.H., Sahakian, B.J., Pickard,J.D., 2005. Head circumference and brain and hippocampal volume after severetraumatic brain injury in childhood. Pediatric Research 58, 302–308.

Teasdale, G., Jennett, B., 1974. Assessment of coma and impaired consciousness.A practical scale. Lancet 2, 81–84.

Verger, K., Junque, C., Levin, H.S., Jurado, M.A., Perez-Gomez, M., Bartres-Faz, D., Barrios,M., Alvarez, A., Bartumeus, F., Mercader, J.M., 2001. Correlation of atrophy measureson MRI with neuropsychological sequelae in children and adolescents withtraumatic brain injury. Brain Injury 15, 211–221.

T. Silk et al. / Psychiatry Research: Neuroimaging 213 (2013) 186–192192

Warner, M.A., Youn, T.S., Davis, T., Chandra, A., Marquez de la Plata, C., Moore, C.,Harper, C., Madden, C.J., Spence, J., McColl, R., Devous, M., King, R.D., Diaz-Arrastia, R., Warner, M.A., Youn, T.S., Davis, T., Chandra, A., Marquez de la Plata,C., Moore, C., Harper, C., Madden, C.J., Spence, J., McColl, R., Devous, M., King, R.D., Diaz-Arrastia, R., 2010. Regionally selective atrophy after traumatic axonalinjury. Archives of Neurology 67, 1336–1344.

Wechsler, D., 1999. Wechsler Abbreviated Scale of Intelligence. The PsychologicalCorporation: Harcourt Brace & Company, New York, NY.

Wilde, E.A., Bigler, E.D., Hunter, J.V., Fearing, M.A., Scheibel, R.S., Newsome, M.R.,Johnson, J.L., Bachevalier, J., Li, X., Levin, H.S., 2007. Hippocampus, amygdala, andbasal ganglia morphometrics in children after moderate-to-severe traumatic braininjury. Developmental Medicine and Child Neurology 49, 294–299.

Wilde, E.A., Hunter, J.V., Newsome, M.R., Scheibel, R.S., Bigler, E.D., Johnson, J.L.,Fearing, M.A., Cleavinger, H.B., Li, X., Swank, P.R., Pedroza, C., Roberson, G.S.,Bachevalier, J., Levin, H.S., 2005a. Frontal and temporal morphometric findingson MRI in children after moderate to severe traumatic brain injury. Journal ofNeurotrauma 22, 333–344.

Wilde, E.A., Hunter, J.V., Newsome, M.R., Scheibel, R.S., Bigler, E.D., Johnson, J.L.,Fearing, M.A., Cleavinger, H.B., Li, X., Swank, P.R., Pedroza, C., Roberson, G.S.,Bachevalier, J., Levin, H.S., Wilde, E.A., Hunter, J.V., Newsome, M.R., Scheibel, R.S., Bigler, E.D., Johnson, J.L., Fearing, M.A., Cleavinger, H.B., Li, X., Swank, P.R.,Pedroza, C., Roberson, G.S., Bachevalier, J., Levin, H.S., 2005b. Frontal andtemporal morphometric findings on MRI in children after moderate to severetraumatic brain injury. Journal of Neurotrauma 22, 333–344.