Implications of reduced callosal area for social skills after severe traumatic brain injury in children

Original Article

Implications of Reduced Callosal Area for Social Skillsafter Severe Traumatic Brain Injury in Children

Miriam H. Beauchamp,1,2,3 Vicki A. Anderson,1,2,3 Cathy Catroppa,1,2,3 Jerome J. Maller,4

Celia Godfrey,1,3 Jeffery V. Rosenfeld,5 and Michael Kean1,3

Abstract

The corpus callosum has been shown to be particularly vulnerable to the effects of traumatic brain injury (TBI) inchildhood, and severe injury frequently results in a smaller corpus callosum post-injury. However, the long-termeffects of TBI on the integrity of the callosum, as well as the potential functional significance of callosal injury arepoorly understood. Some studies suggest the corpus callosum may be involved in social skills, which are oftenreduced following TBI. In this study, callosal size was investigated in 37 individuals 10 years post–childhoodTBI, and its relationship with social competence was examined. The results indicate that individuals who sustainsevere TBI in childhood display reduced callosal size 10 years post-injury. In addition, callosal size correlatedsignificantly with social skills, suggesting that callosal atrophy may be related to poorer social skills after TBI.These findings highlight the persistence of callosal abnormalities, which may be a result of interrupted neuraldevelopment after childhood TBI. The results further underscore the potential importance of the corpus callosumfor social competence after TBI.

Key words: behavior; magnetic resonance imaging; pediatric brain injury; recovery; traumatic brain injury

Introduction

Traumatic brain injury (TBI) in children results inboth structural and functional changes in the brain,

which are associated with cognitive and behavioral deficits(Anderson et al., 2005; Serra-Grabulosa et al., 2005; Wildeet al., 2005). In adults with TBI, there is substantial evidencethat the corpus callosum (CC) is affected by both focal anddiffuse injuries, as reflected by reductions in size ( Johnsonet al., 1996; Tomaiuolo et al., 2004; Yount et al., 2002), and infractional anisotropy, a measure of white matter integrityobserved using diffusion tensor imaging (Sidaros et al., 2008).In children, however, the effect of TBI on the CC remainsuncertain. Children who sustain TBI have been reported tohave callosal lesions (Benavidez et al., 1999; Mendelsohn et al.,1992), and reductions in CC growth have been reported inchildren with severe injuries between 3 and 36 months post-injury, while the area of the CC increases in those with mildand moderate injuries (Levin et al., 2000). Recently, investi-gators have used diffusion tensor imaging (DTI) to examinecallosal pathology post–child-TBI. While some researchers

have reported that white matter integrity is abnormal in theCC (Ewing-Cobbs et al., 2008; Levin et al., 2008; Wilde et al.,2006; Yuan et al., 2007), others have failed to find any changein fractional anisotropy in this area (Wozniak et al., 2007).Discrepancies in results are likely due to a number of factors,including variations in analytical techniques, timing post-injury, injury severity, and age of the patients studied.

Callosal damage has important functional consequencesand has been linked to deficits in interhemispheric transfer ofinformation (Benavidez et al., 1999). Callosal area has alsobeen strongly correlated with processing speed and visuo-spatial function in children with TBI (Verger et al., 2001). Bycontrast, Mathias and colleagues (2004b) were unable toconfirm a relationship between callosal area and interhemi-spheric processing in adults with TBI, despite the presence ofsmaller callosal size and deficits in processing speed. In otherclinical populations (e.g., autism, schizophrenia, psycho-pathy, and callosal agenesis), substantial evidence existslinking an abnormal CC with social problems (Badaruddinet al., 2007; Brambilla et al., 2003; Raine et al., 2003), raising thepossibility that structural changes in this area may also impact

1Australian Centre for Child Neuropsychological Studies, Murdoch Children’s Research Institute, Melbourne, Victoria, Australia.2School of Behavioural Sciences, University of Melbourne, Melbourne, Victoria, Australia.3Royal Children’s Hospital, Melbourne, Victoria, Australia.4Alfred Psychiatry Research Centre, Monash University, Melbourne, Victoria, Australia.5Department of Surgery, Monash University, and Department of Neurosurgery, The Alfred Hospital, Melbourne, Victoria, Australia.

JOURNAL OF NEUROTRAUMA 26:1645–1654 (October 2009)ª Mary Ann Liebert, Inc.DOI: 10.1089=neu.2009.0916

1645

social skills in individuals with TBI. Social skills are deter-mined by the ability to (a) accurately select relevant and usefulinformation from an interpersonal context, (b) use that in-formation to determine appropriate goal-directed behavior,and (c) execute verbal and nonverbal behaviors that maximizethe likelihood of goal attainment and the maintenance of goodrelations with others (Bedell and Lennox, 1997). Social infor-mation processing is a complex ability that requires integra-tion of social problem solving, communication, and affectiveand cognitive abilities, and relies on the efficient transfer ofinformation throughout the brain (Yeates et al., 2007). Inparticular, the relationship between callosal structure andsocial skills has been studied in autism, in an attempt to de-termine a biological basis for the impairments in social be-havior observed in those with this disorder (Hill and Frith,2003). Brain underconnectivity, as defined by reduced size ofthe CC, has been shown to be associated with social impair-ments in autism (Alexander et al., 2007). These findings sup-port the hypothesis that reductions in the interhemisphericconnectivity of the brain may impair efficient informationprocessing of complex stimuli that require coordination be-tween the two hemispheres of the brain, such as that requiredwhen faced with new and intricate social interactions (Bel-monte et al., 2004; Geschwind and Levitt, 2007). In support ofthis, in a recent study researchers found a marginal relation-ship between performance on a social problem-solving taskand lesions of the CC in children with TBI (Hanten et al.,2008).

Several studies to date have focused on global structuralchanges after childhood TBI (Berryhill et al., 1995; Bigler, 1999;Tasker, 2006; Tasker et al., 2005; Verger et al., 2001; Wilde et al.,2005), but comparatively few have investigated the long-termeffects on specific structures. Despite evidence that childrenwho sustain TBI are vulnerable to callosal injury, and thatthey have deficits in social skills ( Janusz et al., 2002; Yeateset al., 2007), the nature of callosal injury following childhoodTBI is unclear, as are its links with social function. Further-more, no study to date has investigated the integrity of the CCin the very long-term after childhood TBI. In this study, theCC was identified as a region of interest and morphometricmeasurements were obtained with the aim of investigatingthe links between callosal area and social function at 10 yearspost-injury. It was expected that reduced callosal area wouldbe associated with poorer social skills in the long-term, par-ticularly in severely injured children.

Methods

Participants

Participants in this study were followed as part of a pro-spective, longitudinal study investigating outcomes of youngchildren with TBI. Since recruitment, these children have beenreviewed acutely, and at 6, 12, 18, and 30 months and 5 yearspost-injury (Anderson and Catroppa, 2005, 2007; Andersonet al., 2000, 2004; Catroppa and Anderson, 2003). This studyreports the data seen at 10-year follow-up. Informed consentwas obtained in compliance with approved ethical guidelinesof the Royal Children’s Hospital, and in accordance with theDeclaration of Helsinki. Participants were assessed at the timeof injury at the neurosurgical ward of the Royal Children’sHospital, Melbourne, Australia. Inclusion criteria were: (1)age at injury 1–7 years; (2) documented evidence of a TBI

(only closed-head injuries were included), as defined below;(3) evidence of a period of altered consciousness; and (4)all subjects were English speakers. Exclusion criteria were:(1) documented history of previous head injury; (2) non-accidental injury; and (3) evidence of a pre-existing physical,neurological, psychiatric, developmental, or attentional dis-order. For the purposes of the current study, participants alsohad to have undergone a magnetic resonance imaging (MRI)scan. No patients had obvious focal lesions or infarction ofthe CC that could have led to ventricular dilation or callosalatrophy different from that caused by diffuse injury.

The sample of patients meeting inclusion criteria for thisstudy comprised 37 individuals. The participants were di-vided into two groups (mild=moderate injuries and severeinjuries), in light of evidence that TBI affects the CC differentlyin severely injured patients compared to those with mild andmoderate injuries (Levin et al., 2000). The categorization ofinjury severity was based on the following criteria: (1)mild=moderate TBI (n¼ 25): Glasgow Coma Scale score(GCS) (Teasdale and Jennett, 1974) on admission 9–15, with atleast some alteration of consciousness level (e.g., drowsinessor disorientation), reduced responsiveness, and=or mass le-sion or other evidence of specific injury on CT=MRI, and=orneurological impairment; and (2) severe TBI (n¼ 12): GCS onadmission 3–8 representing coma, and mass lesion or otherevidence of specific injury on CT=MRI and=or neurologicalimpairment. The demographic and injury characteristics forthe TBI groups are presented in Tables 1 and 2.

Magnetic resonance imaging

Prior to providing consent to undergo MRI, all childrenwere exposed to a mock MRI scanner and given the oppor-tunity to become familiarized with the equipment and pro-cedure. All MR images were acquired with a 1.5T Avantoscanner (Siemens Medical Systems, Erlangen, Germany) atthe Royal Children’s Hospital in Melbourne, Australia. Afterscout scans to align the anterior and posterior commissures, asequence was acquired for manual CC and intracranial area(ICA) estimation; this was a T1-weighted sagittal SPGR ac-quisition (TR¼ 1920, TE¼ 3.93, NEX¼ 1) with 1.0-mm con-tiguous slices. A T2-weighted TSE axial acquisition of slices4 mm thick (TR¼ 4850, TE¼ 103, NEX¼ 2) was also acquired.

Callosal segmentation

For each subject, the midsagittal slice of the T1-weightedsequence was used to obtain an area measurement of the CCand ICA. One investigator, who was blind to the participants’sociodemographic and neuropsychological status, outlinedthe CC area on a midline sagittal view of a T1-weighted sliceof an MRI image using Analyze 6.0� (Brain Imaging Re-source; Mayo Clinic, Rochester, NY). The midline sagittal slicewas chosen using anatomical landmarks in a hierarchicalorder as suggested by Talairach and Tournoux (1993). Theselandmarks were: (1) no or only minimal white matter in thecortical mantle surrounding the CC; (2) the interthalamicadhesion; and (3) the transparent septum pellucidum and avisible aqueduct of Sylvius. The boundaries of the CC wereoutlined using a cordless infrared light-driven cursor, and thearea within the line was calculated (Fig. 1A). Using landmarksadapted from Witelson (1989), the manually segmented CCoutlines were then automatically divided into three sub-

1646 BEAUCHAMP ET AL.

regions. That is, the traced CC was then divided into threeequal-length parallel horizontal divisions by using the ‘‘Di-vision’’ feature in the ROI module of Analyze� software, andthe area within each division was calculated. For these divi-sions, the ‘‘grid’’ option was used to divide the object (theentire CC) vertically into three equal partitions (CC-I, CC-II,and CC-III), which were saved as anterior, midbody, andposterior subregions, respectively, and their areas calculated.Figure 1B shows the partitioning scheme used.

Intra-rater reliability (area 1 – area 2 = [area 1þ area 2] = 2),was tested in 10 randomly selected brains in which the CCarea was traced twice by an experienced manual tracer on

different dates, and was 0.98. Inter-rater reliability, as deter-mined over 10 randomly selected brains, was 0.96.

The ICA was traced from a mid-sagittal slice by drawing aline of the contour of the inner calvarium (Xu et al., 2000). Thelower boundary was a line between the basion and opisthion.Intra-rater reliability over 10 randomly selected scans was0.99, and inter-rater reliability was 0.97.

Behavioral measures

Intellectual function. Full-scale intellectual quotient(FSIQ) was measured acutely using the Bayley Scales of InfantDevelopment (Bayley, 1993), the Wechsler Preschool andPrimary Scale of Intelligence (Wechsler, 1989), or the WechslerIntelligence Scale for Children–R (Wechsler, 1974), and againat 10 years post-injury using the either the Wechsler In-telligence Scale for Children–III (Wechsler, 1991) or theWechsler Adult Intelligence Scale–Third Edition (Wechsler,1997a).

Adaptive Behavior Assessment System (ABAS-II). TheABAS-II (Harrison and Oakland, 2003) is a standardizedquestionnaire assessing adaptive, functional, and practicalskills. The social subscale score of the parent report form wasused to measure adaptive social functioning and social com-petence in this study. The scale relates to the skills needed tointeract socially and get along with other people, includingexpressing affection, having friends, showing and recogniz-ing emotions, assisting others, and using manners (scaledscore; mean¼ 10, SD¼ 3).

Behavior assessment system for children (BASC). TheBASC (Reynolds and Kamphaus, 1992) is a standardizedquestionnaire assessing various domains of behavior. Thesocial skills adaptive scale was used from the parent reportform (T score; mean¼ 50, SD¼ 10).

Social Skills Rating Scale (SSRS). The SSRS (Greshamand Elliott, 1990) is a standardized parent report designed toobtain information on social and problem behaviors. The so-cial skills scale was used to measure the level of positive socialbehaviors. The total social scale includes four subscales: thecooperation subscale includes behaviors such as helpingothers and complying with rules and directions; the asser-tion subscale includes initiating social behaviors such as in-troducing oneself and responding to the actions of others; theresponsibility subscale includes behaviors related to com-munication with adults and regard for the property of others;

Table 1. Demographic and Injury Characteristics

for Patients with TBI

Mild tomoderate Severe

Demographic dataTotal number 25 12Male, number (%) 15 (60) 8 (67)Left-handed, number (%) 5 (20) 1 (8)Family socioeconomic

status,a mean (SD)4.24 (0.75) 4.08 (1.05)

Age at acute assessment(years), mean (SD)

4.99 (1.92) 4.85 (1.91)

Age at 10-year follow-up(years), mean (SD)

14.90 (2.88) 15.60 (2.82)

Time since injury,mean (SD)

10.12 (1.59) 11.07 (1.44)

FSIQ acute, mean (SD) 100.55 (16.29) 90.77 (23.36)FSIQ at 10 years,

mean (SD)100.63 (16.45) 91.85 (17.32)

Injury characteristicsGCS on admission,

mean (SD)c11.28 (3.65) 5.50 (2.11)

Neurological signs,number (%)c

7 (28) 11 (92)

Abnormal CT=MRI,number (%)c

15 (60) 12 (100)

Surgical intervention,number (%)c

6 (24) 8 (67)

Comac

No coma, number (%) 16 (64) —LOC 1 sec–5 min, number (%) 5 (20) —LOC 0.5 day, number (%) 3 (12) 2 (17)LOC 1–3 days, number (%) 1 (4)b 2 (17)LOC 4–7 days, number (%) — 2 (17)LOC 8–14 days, number (%) — 5 (42)LOC> 14 days, number (%) — 1 (8)

Cause of injuryc

Passenger in auto accident,number (%)

1 (4) 2 (17)

Pedestrian in auto accident,number (%)

2 (8) 4 (33)

Fall, number (%) 20 (80) 3 (25)Other, number (%) 2 (8) 3 (25)

aDaniel’s Scale of Occupational Prestige (Daniel, 1983): low scorereflects high occupational prestige (range: 1.0–6.9).

bLength of coma (LOC) secondary to other injuries and unrelatedto TBI severity.

cStatistically significant ( p< 0.05).FSIQ, full-scale intellectual quotient; GCS, Glasgow Coma Scale

score.

Table 2. Initial Location of Pathology

Measured on CT

Pathology locationMild to

moderate Severe

None, number (%) 14 (70) 2 (18)Frontal, number (%) 3 (15) 4 (36)Temporal, number (%) 1 (5) 1 (9)Parietal, number (%) 1 (5) —Frontal and extrafrontal, number (%) — 2 (18)Subcortical, number (%) — 1 (9)Cortical and subcortical, number (%) — 1 (9)Combined extrafrontal, number (%) 1 (5) —

CALLOSAL AREA AND SOCIAL SKILLS AFTER CHILD TBI 1647

FIG. 1. Example of manual segmentation of the mid-sagittal corpus callosum area. (A) Total mid-sagittal area. (B) Totalmid-sagittal corpus callosum area divided into three equal segments.

and the self-control subscale includes behaviors that emergein conflict situations, such as responding appropriately toteasing, and non-conflict situations, such as turn-taking(standard score; mean¼ 100, SD¼ 15).

Processing speed index (PSI). The PSI was used as ameasure of the speed of information processing in order todetermine the efficiency of interhemispheric transfer, and wasderived from either the Wechsler Adult Scale of Intelligence(WAIS-III) (Wechsler, 1997b) or the Wechsler IntelligenceScale for Children (WISC-III) (Wechsler, 1991), depending onthe age of the subject (standard score; mean¼ 100, SD¼ 15).

Statistical analysis

Differences in demographic and injury characteristics wereassessed between the two patient groups using independentsample t-tests for continuous variables, and Pearson’s chi-square statistic for categorical variables. Analysis of covariance(ANCOVA) was used to examine group differences in callosalarea, adjusting for intracranial area and age at MRI. In-dependent sample t-tests were used to compare the results ofthe mild=moderate and severe TBI groups on measures of so-cial skills and processing speed. In addition, the percentage ofparticipants with a borderline or clinically significant deficit(defined as a mean score of >1.5 SD from the mean or aT-score>60) was also reported. Comparisons of percentages ofindividuals with deficits in each severity group involvedPearson’s chi-square statistic. Bivariate correlations were per-formed between the processing speed index and the socialvariables (ABAS-II, BASC, and SSRS) to investigate the rela-tionship between processing speed and social function. Partialcorrelations taking into account age and intracranial area wereused to examine the relationship between callosal area andsocial competence, as well as information processing speed.Intracranial area was used as a covariate to account for overalldifferences in brain size between participants. Although therewere no significant age differences between the participantgroups, age was included as a covariate to account for subtledevelopmental differences (Wilde et al., 2007).

Results

The TBI groups were comparable on all demographiccharacteristics (Table 1). As expected, the two TBI groupsdiffered in injury characteristics (Table 1). ANOVA indicateda significant difference between the two groups on GCS atadmission [F (1,36)¼ 25.76, p¼ 0.001). Pearson’s chi-squarerevealed that there was a significant relationship betweenseverity group and neurological signs [w2 (1,37)¼ 13.16,p¼ 0.001; abnormal CT=MRI, w2 (1,37)¼ 6.58, p¼ 0.01; andneed for surgical intervention, w2 (1,37)¼ 6.28, p¼ 0.01].Pearson’s chi-square also showed a significant relationshipbetween severity and duration of coma [w2 (6,37)¼ 28.48,p¼ 0.001; and cause of injury, w2 (3,37)¼ 20.49, p¼ 0.02]. Nosignificant differences were found with regard to initial lo-cation of pathology (Table 2).

Callosal area

The severe TBI group was found to have significantlysmaller total callosal area compared to the mild=moderatepatients (Table 3 and Fig. 2). The posterior part of the CC in

particular was found to be significantly smaller in the severelyinjured patients.

Behavioral measures

Patient results on the behavioral measures are presented inTable 4. A substantial number (19–41%) of the patients hadclinically significant deficits on measures of social skills andinformation processing. In addition, patients with severe TBIhad significantly worse results on the ABAS-II [t (1,35)¼ 2.4,p¼ 0.02] and PSI [t (1,35)¼ 1.9, p¼ 0.05] than those with mildto moderate injuries. Correlation analyses revealed a signifi-cant positive association between processing speed and socialfunction as assessed by the ABAS-II (r¼ 0.53, p¼ 0.001) andthe SSRS assertion scale (r¼ 0.42, p¼ 0.01), indicating thatfaster processing speeds were related to better social skills onthese measures.

Relationship between callosal area and social skills

When age and intracranial area were taken into account,processing speed and social skills, as measured by the ABAS-II, BASC, and PSI, correlated significantly with callosal area,indicating that smaller callosal area was associated withslower processing speed and poorer social skills (Table 5).This relationship was particularly significant for the middleand posterior portions of the CC. The SSRS did not correlatesignificantly overall with callosal area; however, a near-significant correlation was found between the SSRS assertionsubscale and total mid-sagittal area (r¼ 0.31, p¼ 0.08), as wellas the middle portion of the CC (r¼ 0.33, p¼ 0.06).

Discussion

As predicted, the patients in this study with severe child-hood TBI were found to have significantly smaller corporacallosa than mild- and moderately-injured patients. To ourknowledge, this is the first evidence of persisting callosalatrophy in a longitudinal sample, all of whom were injured 10years prior, indicating lasting brain changes in this area, andhighlighting the long-term vulnerability of the region. Pre-viously, Levin and colleagues (2000) reported a decrease incallosal size between 3 and 36 months post-TBI in severelyinjured children. This was in contrast to their mild=moderategroup of patients, who demonstrated an increase in callosalsize during the same period of time. Smaller callosal size hasalso been observed in severely-injured adults compared tomild- and moderately-injured patients and healthy controlparticipants, though these studies were not conducted overthe long-term post-injury period ( Johnson et al., 1996; Levinet al., 1990; Tomaiuolo et al., 2004; Yount et al., 2002).

Table 3. Area of the Corpus Callosum in TBI Patients

Brain area (mm2)Total

(n¼ 37)

Mild tomoderate(n¼ 25)

Severe(n¼ 12) Fa p

Total mid-sagittal 618 (84) 647 (84) 558 (156) 4.5 .04Anterior CC 243 (49) 251 (39) 226 (64) 1.4 .24Middle CC 140 (30) 144 (24) 130 (39) 1.3 .25Posterior CC 237 (55) 251 (41) 208 (70) 5.0 .03

aCovarying for age and intracranial area.

CALLOSAL AREA AND SOCIAL SKILLS AFTER CHILD TBI 1649

In this study, patients were evaluated 10 years post-injury,and the findings suggest that there is no recovery of callosalsize in the long-term after severe TBI. It is likely that suchlasting effects are a result of sustaining TBI at a crucial time indevelopment. Studies of CC growth indicate that the structurecontinues to develop into the third decade of life, in parallelwith maturation of higher-association areas throughout ado-lescence and early adulthood (Giedd et al., 1996; Pujol et al.,1993). Given that the development of callosal axons is com-plete before birth, Giedd and colleagues (1996) suggest thatmyelination of these axons is directly related to the increase incallosal size observed throughout childhood and adolescence.Thus childhood TBI may disrupt the growth of this area in anumber of interactive and additive ways: (1) through acutediffuse axonal injury and shearing of the white matter tractscausing atrophy ( Johnson et al., 1996); (2) by causing pro-

gressive neural degeneration and shrinkage of the CC withtime (Levin et al., 2000; Tomaiuolo et al., 2004; Verger et al.,2001); or (3) by interfering with developmental myelination ofthe CC.

The posterior third of the CC, comprising the isthmus andsplenium, was particularly affected by severe injury in thecurrent study. This is consistent with findings that the posteriorregion is the most frequent site of injury following TBI (Bena-videz et al., 1999; Levin et al., 2000; Povlishock, 1992; Pov-lishock et al., 1992). This region may be additionally susceptibleto the effects of childhood TBI because maturational changesare primarily observed in the posterior section, whereas ante-rior regions likely reach adult size in early childhood (Gieddet al., 1996). These findings are also supported by evidencefrom DTI that callosal growth in posterior regions is arrested inyouth with TBI (Ewing-Cobbs et al., 2008).

Table 4. Average Scores and Percentage of TBI Patients with a Deficit on Behavioral

Measures at 10 Years Post-Injury

Measure Total TBI (n¼ 37) Mild=moderate (n¼ 25) Severe (n¼ 12)

SSRS total, mean (SD) 100.9 (18.6) 99.9 (19.9) 102.9 (16.5)SSRS deficit, (%) 19 21 17BASC social scale, mean (SD) 49.2 (11.5) 49.3 (12.9) 49.0 (9.0)BASC deficit, (%) 41 45 50ABAS-II social scale, mean (SD)a 8.5 (3.8) 9.5 (3.4) 6.3 (3.8)ABAS-II deficit, (%)a 29 17 60PSI, mean (SD)a 96 (18.8) 100 (16) 88 (22)PSI deficit, (%) 27 20 42

ap< 0.05.ABAS, Adaptive Behavior Assessment System-II (scaled score); BASC, behavior assessment system for children (T score); PSI, processing

speed index (standard score); SSRS, Social Skills Rating Scale (standard score).

FIG. 2. Callosal area (mm2) in patients with TBI at 10 years post-injury.

1650 BEAUCHAMP ET AL.

The functional significance of CC atrophy after TBI has beenalluded to in previous studies. Findings from both volumetricand DTI studies point to links between callosal integrity andinterhemispheric functions such as dichotic listening (Bena-videz et al., 1999), speed of processing (Levin et al., 2008;Verger et al., 2001; Wilde et al., 2006), and visuospatial func-tion (Verger et al., 2001). Considerable evidence exists sup-porting a topographical organization of the CC, suggestingposterior portions, such as the splenium, contain larger axonsconnecting visual and somatosensory cortices (Aboitiz et al.,1992), and link with posterior regions of the brain. Accordingto Witelson (1989), the isthmus maintains connections with thesuperior temporal and posterior parietal region, while thesplenium carries occipital and inferior temporal fibers. How-ever, Doron and Gazzaniga (2008) highlight the complexity ofstructural, functional, and connective targets of the CC, andsuggest there may be exceptions to the topographical ar-rangement, as well as substantial individual variation.

The results presented here indicate that 10 years post-injury, many TBI patients, particularly those with severe in-juries, have poor social skills and slow speeds of processing.This is consistent with an emerging body of evidence thatindicates that the long-term consequences of childhood TBImay extend beyond cognitive dysfunction to social and be-havioral domains (Ganesalingam et al., 2006; Hanten et al.,2008; Janusz et al., 2002; Kirkwood et al., 2000; Yeates et al.,2004) and to slower processing (Catroppa et al., 2007; Mathiaset al., 2004a). In our study, performance on measures of socialskills and processing speed correlated with callosal area, in-dicating a significant relationship between a smaller CC andpoorer social skills, as well as slower speeds of processing.Researchers have postulated more generally that the CC mayplay an important role in the development and maintenance ofgood social skills. This idea stems from its global role in theintegration of activity and information between the twohemispheres, and transfer of important sensory and higher-order stimuli. Much of the evidence for this has come fromother populations with known callosal abnormalities, such asthose with autism and callosal agenesis, as both groups havesignificant deficits in social interaction, recognition of emo-tions, comprehending humor and socially complex scenarios,detecting changes in prosody, and in social intelligence moregenerally (Badaruddin et al., 2007; Boger-Megiddo et al., 2006;Brambilla et al., 2003; Brown et al., 2005; Hrdlicka, 2008; Paulet al., 2007; Paul et al., 2006; Paul et al., 2004; Schmitz andRezaie, 2008). In addition, a recent study in children with TBIindicated a marginal relationship between the presence ofcallosal lesions and poor performance on a social problem-solving task (Hanten et al., 2008).

Studies in healthy individuals support the idea that inter-hemispheric communication is important for the socio-emotional abilities that underlie social competence(Wildgruber et al., 2005). For example, studies investigatingthe rapid recognition of social emotion indicate that inter-hemispheric cooperation and interaction are necessary for thisability (Tamietto et al., 2007; Tamietto et al., 2006). In thecurrent study a significant relationship was found betweentwo general measures of social skills, as well as informationprocessing. In addition, faster information processing speedswere significantly related to better social function. It is likelythat these links represent an indirect role for the CC in socialskills that rely on complex analysis of rapidly changing in-formation and fast processing.

This study had some methodological limitations. First, theabsence of a control group precluded comparison of CC areain patients with TBI to that of normally-developing individ-uals. Previous studies including control groups have shownthat CC area is reduced in patients with TBI (e.g., Tomaiuoloet al., 2004); however, it is not clear whether this remains trueat 10 years post-childhood TBI. Also, it would be ideal toinvestigate the change in CC structure longitudinally in thissample; however, comparable brain imaging data were notavailable at the initial injury time-point. Nonetheless, the re-sults showing the differences in children with varying degreesof injury severity provide important information on the im-pact of TBI on the development and integrity of the CC in thelong term. Second, the behavioral measures of social skillsused in this study were limited to parental reports of socialfunction. Future studies would benefit from using more directand specific measures of social skills in order to preciselydefine the nature of the social deficits observed following TBI,as well as its relationship to callosal area. Third, in this study,the CC was divided into three sections. In some other studiesresearchers have divided this structure into seven parts (e.g.,Giedd et al., 1996), which may result in greater specificity indetermining the exact portions of the CC associated withparticular skills. Fourth, the data presented here relate spe-cifically to the integrity of the CC post-TBI; analysis of lesionsin other regions associated with cognitive and social dys-function may also be of benefit in the future, and may con-tribute to our understanding of the overall impact of regionalstructural changes on social functioning over the long termpost-injury.

Conclusions

In summary, the findings of this study confirm the long-term persistence of callosal atrophy in individuals who sus-tained severe TBI during childhood. These results are thoughtto be due to developmental disruption of callosal growthduring adolescence. Furthermore, the results highlight a keylink between reduced callosal size and poor social skills, aswell as slower processing speeds at 10 years post-injury. It islikely that the CC plays an integral role in social skills byfacilitating the rapid transfer and coordination of importantsocial information between the two hemispheres. Althoughwe cannot infer causation from the correlations reported here,the significant link detected between social skills and callosalarea is notable and warrants further investigation using moresensitive and precise measures of social functioning, as well asDTI and functional brain imaging methods.

Table 5. Partial Correlations Between Social Skills

and CC area, r (p)

PSIBASCsocial

ABAS-IIsocial

SSRSassertion

Total mid-sagittalarea

.35 (.04) .50 (.005) .55 (.001) .31 (.08)

Anterior CC .25 (.15) .27 (.15) .33 (.07) .27 (.13)Middle CC .33 (.06) .42 (.02) .49 (.005) .33 (.06)Posterior CC .32 (.06) .49 (.007) .58 (<.001) .12 (.49)

CALLOSAL AREA AND SOCIAL SKILLS AFTER CHILD TBI 1651

Acknowledgment

This work was supported by the National Health andMedical Research Council (project grant no. 284518).

Author Disclosure Statement

No conflicting financial interests exist.

References

Aboitiz, F., Scheibel, A.B., Fisher, R.S., and Zaidel, E. (1992).Fiber composition of the human corpus callosum. Brain Res.598, 143–153.

Alexander, A.L., Lee, J.E., Lazar, M., Boudos, R., DuBray, M.B.,Oakes, T.R., Miller, J.N., Lu, J., Jeong, E.K., McMahon, W.M.,Bigler, E.D., and Lainhart, J.E. (2007). Diffusion tensor imagingof the corpus callosum in autism. NeuroImage 34, 61–73.

Anderson, V.A., Catroppa, C., Rosenfeld, J., Haritou, F., andMorse, S.A. (2000). Recovery of memory function followingtraumatic brain injury in pre-school children. Brain Inj. 14,679–692.

Anderson, V., and Catroppa, C. (2007). Memory outcome at 5years post-childhood traumatic brain injury. Brain Inj. 21,1399–1409.

Anderson, V., and Catroppa, C. (2005). Recovery of executiveskills following paediatric traumatic brain injury (TBI): a 2year follow-up. Brain Inj. 19, 459–470.

Anderson, V., Catroppa, C., Morse, S., Haritou, F., and Ro-senfeld, J. (2005). Functional plasticity or vulnerability afterearly brain injury? Pediatrics 116, 1374–1382.

Anderson, V.A., Morse, S.A., Catroppa, C., Haritou, F., and Ro-senfeld, J.V. (2004). Thirty month outcome from early childhoodhead injury: a prospective analysis of neurobehavioural recov-ery. Brain 127, 2608–2620.

Badaruddin, D.H., Andrews, G.L., Bolte, S., Schilmoeller, K.J.,Schilmoeller, G., Paul, L.K., and Brown, W.S. (2007). Socialand behavioral problems of children with agenesis of thecorpus callosum. Child Psychiatry Hum. Dev. 38, 287–302.

Bayley, N. (1993). Bayley Scales of Infant Development—SecondEdition. The Psychological Corporation: San Antonio, TX.

Bedell, J.R., and Lennox, S.S. (1997). Handbook for Communicationand Problem-solving Skills Training: A Cognitive-BehaviouralApproach. John Wiley and Sons: New York.

Belmonte, M.K., Allen, G., Beckel-Mitchener, A., Boulanger,L.M., Carper, R.A., and Webb, S.J. (2004). Autism and ab-normal development of brain connectivity. J. Neurosci. 24,9228–9231.

Benavidez, D.A., Fletcher, J.M., Hannay, H.J., Bland, S.T., Caudle,S.E., Mendelsohn, D.B., Yeakley, J., Brunder, D.G., Harward,H., Song, J., Perachio, N.A., Bruce, D., Scheibel, R.S., Lilly, M.A.,Verger-Maestre, K., and Levin, H.S. (1999). Corpus callosumdamage and interhemispheric transfer of information follow-ing closed head injury in children. Cortex 35, 315–336.

Berryhill, P., Lilly, M.A., Levin, H.S., Hillman, G.R., Mendel-sohn, D., Brunder, D.G., Fletcher, J.M., Kufera, J., Kent, T.A.,Yeakley, J., Bruce, D., and Eisenberg, H.M. (1995). Frontal lobechanges after severe diffuse closed head injury in children: avolumetric study of magnetic resonance imaging. Neuro-surgery 37, 392–399; discussion 399–400.

Bigler, E.D. (1999). Neuroimaging in pediatric traumatic headinjury: diagnostic considerations and relationships to neuro-behavioral outcome. J. Head Trauma Rehabil. 14, 406–423.

Boger-Megiddo, I., Shaw, D.W., Friedman, S.D., Sparks, B.F.,Artru, A.A., Giedd, J.N., Dawson, G., and Dager, S.R. (2006).

Corpus callosum morphometrics in young children with au-tism spectrum disorder. J. Autism Dev. Disord. 36, 733–739.

Brambilla, P., Hardan, A., di Nemi, S.U., Perez, J., Soares, J.C.,and Barale, F. (2003). Brain anatomy and development inautism: review of structural MRI studies. Brain Res. Bull. 61,557–569.

Brown, W.S., Paul, L.K., Symington, M., and Dietrich, R. (2005).Comprehension of humor in primary agenesis of the corpuscallosum. Neuropsychologia 43, 906–916.

Catroppa, C., and Anderson, V. (2003). Children’s attentionalskills 2 years post-traumatic brain injury. Dev. Neuropsychol.23, 359–373.

Catroppa, C., Anderson, V.A., Morse, S.A., Haritou, F., and Ro-senfeld, J.V. (2007). Children’s attentional skills 5 years post-TBI. J. Pediatr. Psychol. 32, 354–369.

Daniel, A. (1983). Power, privilege, and prestige: Occupations inAustralia. Melbourne, Australia: Longman-Cheshire.

Doron, K.W., and Gazzaniga, M.S. (2008). Neuroimaging tech-niques offer new perspectives on callosal transfer and inter-hemispheric communication. Cortex 44, 1023–1029.

Ewing-Cobbs, L., Prasad, M.R., Swank, P., Kramer, L., Cox, C.S.,Jr., Fletcher, J.M., Barnes, M., Zhang, X., and Hasan, K.M.(2008). Arrested development and disrupted callosal micro-structure following pediatric traumatic brain injury: relation toneurobehavioral outcomes. NeuroImage 42, 1305–1315.

Ganesalingam, K., Sanson, A., Anderson, V., and Yeates, K.O.(2006). Self-regulation and social and behavioral functioningfollowing childhood traumatic brain injury. J. Int. Neu-ropsychol. Soc. 12, 609–621.

Geschwind, D.H., and Levitt, P. (2007). Autism spectrum dis-orders: developmental disconnection syndromes. Curr. Opin.Neurobiol. 17, 103–111.

Giedd, J.N., Rumsey, J.M., Castellanos, F.X., Rajapakse, J.C.,Kaysen, D., Vaituzis, A.C., Vauss, Y.C., Hamburger, S.D., andRapoport, J.L. (1996). A quantitative MRI study of the corpuscallosum in children and adolescents. Brain Res. Dev. BrainRes. 91, 274–280.

Gresham, F.M., and Elliott, S.N. (1990). Social Skills Rating Sys-tem. American Guidance Service: Circle Pines, MN.

Hanten, G., Wilde, E.A., Menefee, D.S., Li, X., Lane, S., Vasquez,C., Chu, Z., Ramos, M.A., Yallampalli, R., Swank, P., Chap-man, S.B., Gamino, J., Hunter, J.V., and Levin, H.S. (2008).Correlates of social problem solving during the first year aftertraumatic brain injury in children. Neuropsychology 22, 357–370.

Harrison, P.L., and Oakland, T. (2003). Adaptive Behavior Assess-ment System—Second Edition. The Psychological Corporation:San Antonio, TX.

Hill, E.L., and Frith, U. (2003). Understanding autism: insightsfrom mind and brain. Philos. Trans. R. Soc. Lond. B. Biol. Sci.358, 281–289.

Hrdlicka, M. (2008). Structural neuroimaging in autism. Neuro.Endocrinol. Lett. 29, 281–286.

Janusz, J.A., Kirkwood, M.W., Yeates, K.O., and Taylor, H.G.(2002). Social problem-solving skills in children with traumaticbrain injury: long-term outcomes and prediction of socialcompetence. Child Neuropsychol. 8, 179–194.

Johnson, S.C., Pinkston, J.B., Bigler, E.D., and Blatter, D.D.(1996). Corpus callosum morphology in normal controls andtraumatic brain injury: Sex differences, mechanisms of injury,and neuropsychological correlates. Neuropsychology 10, 408–415.

Kirkwood, M., Janusz, J., Yeates, K.O., Taylor, H.G., Wade, S.L.,Stancin, T., and Drotar, D. (2000). Prevalence and correlates of

1652 BEAUCHAMP ET AL.

depressive symptoms following traumatic brain injuries inchildren. Child Neuropsychol. 6, 195–208.

Levin, H.S., Benavidez, D.A., Verger-Maestre, K., Perachio, N.,Song, J., Mendelsohn, D.B., and Fletcher, J.M. (2000). Reduc-tion of corpus callosum growth after severe traumatic braininjury in children. Neurology 54, 647–653.

Levin, H.S., Wilde, E.A., Chu, Z., Yallampalli, R., Hanten, G.R.,Li, X., Chia, J., Vasquez, A.C., and Hunter, J.V. (2008). Diffu-sion tensor imaging in relation to cognitive and functionaloutcome of traumatic brain injury in children. J. Head TraumaRehabil. 23, 197–208.

Levin, H.S., Williams, D.H., Valastro, M., Eisenberg, H.M.,Crofford, M.J., and Handel, S.F. (1990). Corpus callosal atro-phy following closed head injury: detection with magneticresonance imaging. J. Neurosurg. 73, 77–81.

Mathias, J.L., Beall, J.A., and Bigler, E.D. (2004a). Neuro-psychological and information processing deficits followingmild traumatic brain injury. J. Int. Neuropsychol. Soc. 10, 286–297.

Mathias, J.L., Bigler, E.D., Jones, N.R., Bowden, S.C., Barrett-Woodbridge, M., Brown, G.C., and Taylor, D.J. (2004b).Neuropsychological and information processing performanceand its relationship to white matter changes following mod-erate and severe traumatic brain injury: a preliminary study.Appl. Neuropsychol. 11, 134–152.

Mendelsohn, D., Levin, H.S., Bruce, D., Lilly, M., Harward, H.,Culhane, K.A., and Eisenberg, H.M. (1992). Late MRI afterhead injury in children: relationship to clinical features andoutcome. Childs Nerv. Syst. 8, 445–452.

Paul, L.K., Brown, W.S., Adolphs, R., Tyszka, J.M., Richards, L.J.,Mukherjee, P., and Sherr, E.H. (2007). Agenesis of the corpuscallosum: genetic, developmental and functional aspects ofconnectivity. Nat. Rev. Neurosci. 8, 287–299.

Paul, L.K., Lautzenhiser, A., Brown, W.S., Hart, A., Neumann,D., Spezio, M., and Adolphs, R. (2006). Emotional arousal inagenesis of the corpus callosum. Int. J. Psychophysiol. 61, 47–56.

Paul, L.K., Schieffer, B., and Brown, W.S. (2004). Social proces-sing deficits in agenesis of the corpus callosum: narrativesfrom the Thematic Appreciation Test. Arch. Clin. Neuro-psychol. 19, 215–225.

Povlishock, J.T., Erb, D.E., and Astruc, J. (1992). Axonal responseto traumatic brain injury: reactive axonal change, deaf-ferentation, and neuroplasticity. J. Neurotrauma 9 (Suppl. 1),S189–S200.

Povlishock, J.T. (1992). Traumatically induced axonal injury:pathogenesis and pathobiological implications. Brain Pathol.2, 1–12.

Pujol, J., Vendrell, P., Junque, C., Marti-Vilalta, J.L., and Cap-devila, A. (1993). When does human brain development end?Evidence of corpus callosum growth up to adulthood. Ann.Neurol. 34, 71–75.

Raine, A., Lencz, T., Taylor, K., Hellige, J.B., Bihrle, S., Lacasse,L., Lee, M., Ishikawa, S., and Colletti, P. (2003). Corpuscallosum abnormalities in psychopathic antisocial individuals.Arch. Gen. Psychiatry 60, 1134–1142.

Reynolds, C.R., and Kamphaus, R.W. (1992). Behavior AssessmentSystem for Children. American Guidance Service: Circle Pines,MN.

Schmitz, C., and Rezaie, P. (2008). The neuropathology of au-tism: where do we stand? Neuropathol. Appl. Neurobiol. 34,4–11.

Serra-Grabulosa, J.M., Junque, C., Verger, K., Salgado-Pineda, P.,Maneru, C., and Mercader, J.M. (2005). Cerebral correlates of

declarative memory dysfunctions in early traumatic brain in-jury. J. Neurol. Neurosurg. Psychiatry 76, 129–131.

Sidaros, A., Engberg, A.W., Sidaros, K., Liptrot, M.G., Herning,M., Petersen, P., Paulson, O.B., Jernigan, T.L., and Rostrup, E.(2008). Diffusion tensor imaging during recovery from severetraumatic brain injury and relation to clinical outcome: alongitudinal study. Brain 131, 559–572.

Talairach, J., and Tournoux, P. (1993). Referentially OrientedCerebral Anatomy: Atlas of Stereotaxic Anatomical Correlations forGray and White Matter. Thieme Medical Publishers, Inc.: NewYork.

Tamietto, M., Adenzato, M., Geminiani, G., and de Gelder, B.(2007). Fast recognition of social emotions takes the wholebrain: interhemispheric cooperation in the absence of cerebralasymmetry. Neuropsychologia 45, 836–843.

Tamietto, M., Latini Corazzini, L., de Gelder, B., and Geminiani,G. (2006). Functional asymmetry and interhemispheric coop-eration in the perception of emotions from facial expressions.Exp. Brain Res. 171, 389–404.

Tasker, R.C. (2006). Changes in white matter late after severetraumatic brain injury in childhood. Dev. Neurosci. 28, 302–308.

Tasker, R.C., Salmond, C.H., Westland, A.G., Pena, A., Gillard,J.H., Sahakian, B.J., and Pickard, J.D. (2005). Head circumfer-ence and brain and hippocampal volume after severe trau-matic brain injury in childhood. Pediatr. Res. 58, 302–308.

Teasdale, G., and Jennett, B. (1974). Assessment of coma andimpaired consciousness. Lancet 2, 81–84.

Tomaiuolo, F., Carlesimo, G.A., Di Paola, M., Petrides, M., Fera,F., Bonanni, R., Formisano, R., Pasqualetti, P., and Caltagir-one, C. (2004). Gross morphology and morphometric sequelaein the hippocampus, fornix, and corpus callosum of patientswith severe non-missile traumatic brain injury without mac-roscopically detectable lesions: a T1 weighted MRI study.J. Neurol. Neurosurg. Psychiatry 75, 1314–1322.

Verger, K., Junque, C., Levin, H.S., Jurado, M.A., Perez-Gomez,M., Bartres-Faz, D., Barrios, M., Alvarez, A., Bartumeus, F.,and Mercader, J.M. (2001). Correlation of atrophy mea-sures on MRI with neuropsychological sequelae in childrenand adolescents with traumatic brain injury. Brain Inj. 15, 211–221.

Wechsler, D. (1974). Manual for the Wechsler Intelligence Scale forChildren—Revised. The Psychological Corporation: New York.

Wechsler, D. (1997a). Wechsler Adult Intelligence Scale—ThirdEdition Manual. The Psychological Corporation: San Antonio,TX.

Wechsler, D. (1997b). Wechsler Adult Intelligence Scale (WAIS-III)Administration and Scoring Manual, 3rd ed. The PsychologicalCorporation: San Antonio, TX.

Wechsler, D. (1991). Wechsler Intelligence Scale for Children—ThirdEdition: Manual, 3rd ed. The Psychological Corporation: SanAntonio, TX.

Wechsler, D. (1989). Wechsler Preschool and Primary Scale ofIntelligence—Revised. The Psychological Corporation: San An-tonio, TX.

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.,and Levin, H.S. (2007). Hippocampus, amygdala, andbasal ganglia morphometrics in children after moderate-to-severe traumatic brain injury. Dev. Med. Child Neurol. 49,294–299.

Wilde, E.A., Chu, Z., Bigler, E.D., Hunter, J.V., Fearing, M.A.,Hanten, G., Newsome, M.R., Scheibel, R.S., Li, X., and Levin,H.S. (2006). Diffusion tensor imaging in the corpus callosum

CALLOSAL AREA AND SOCIAL SKILLS AFTER CHILD TBI 1653

in children after moderate to severe traumatic brain injury.J. Neurotrauma 23, 1412–1426.

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., andLevin, H.S. (2005). Frontal and temporal morphometric find-ings on MRI in children after moderate to severe traumaticbrain injury. J. Neurotrauma 22, 333–344.

Wildgruber, D., Riecker, A., Hertrich, I., Erb, M., Grodd, W.,Ethofer, T., and Ackermann, H. (2005). Identification of emo-tional intonation evaluated by fMRI. NeuroImage 24, 1233–1241.

Witelson, S.F. (1989). Hand and sex differences in the isthmusand genu of the human corpus callosum. A postmortemmorphological study. Brain 112 (Pt. 3), 799–835.

Wozniak, J.R., Krach, L., Ward, E., Mueller, B.A., Muetzel, R.,Schnoebelen, S., Kiragu, A., and Lim, K.O. (2007). Neurocog-nitive and neuroimaging correlates of pediatric traumaticbrain injury: a diffusion tensor imaging (DTI) study. Arch.Clin. Neuropsychol. 22, 555–568.

Xu, Y., Jack, C.R., Jr., O’Brien, P.C., Kokmen, E., Smith, G.E.,Ivnik, R.J., Boeve, B.F., Tangalos, R.G., and Petersen, R.C.(2000). Usefulness of MRI measures of entorhinal cortex ver-sus hippocampus in AD. Neurology 54, 1760–1767.

Yeates, K.O., Bigler, E.D., Dennis, M., Gerhardt, C.A., Rubin,K.H., Stancin, T., Taylor, H.G., and Vannatta, K. (2007). Socialoutcomes in childhood brain disorder: A heuristic integration

of social neuroscience and developmental psychology. Psy-chol. Bull. 133, 535–556.

Yeates, K.O., Swift, E., Taylor, H.G., Wade, S.L., Drotar, D.,Stancin, T., and Minich, N. (2004). Short- and long-term socialoutcomes following pediatric traumatic brain injury. J. Int.Neuropsychol. Soc. 10, 412–426.

Yount, R., Raschke, K.A., Biru, M., Tate, D.F., Miller, M.J.,Abildskov, T., Gandhi, P., Ryser, D., Hopkins, R.O., and Big-ler, E.D. (2002). Traumatic brain injury and atrophy of thecingulate gyrus. J. Neuropsychiatry Clin. Neurosci. 14, 416–423.

Yuan, W., Holland, S.K., Schmithorst, V.J., Walz, N.C., Cecil,K.M., Jones, B.V., Karunanayaka, P., Michaud, L., and Wade,S.L. (2007). Diffusion tensor MR imaging reveals persistentwhite matter alteration after traumatic brain injury experi-enced during early childhood. A.J.N.R. Am. J. Neuroradiol. 28,1919–1925.

Address correspondence to:Miriam H. Beauchamp, Ph.D.

Department of PsychologyRoyal Children’s Hospital

Flemington RoadParkville, Victoria 3052, Australia

E-mail: [email protected]

1654 BEAUCHAMP ET AL.

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.