Superior temporal gyrus volumes in pediatric generalized anxiety disorder

ORIGINAL ARTICLES Superior Temporal Gyrus Volumes in Pediatric Generalized Anxiety Disorder Michael D. De Bellis, Matcheri S. Keshavan, Heather Shifflett, Satish Iyengar, Ronald E. Dahl, David A. Axelson, Boris Birmaher, Julie Hall, Grace Moritz, and Neal D. Ryan Background: The essential symptoms of generalized anxiety disorder (GAD) are intrusive worry about everyday life circumstances and social competence, and associated autonomic hyperarousal. The amygdala, a brain region involved in fear and fear-related behaviors in animals, and its projections to the superior temporal gyrus (STG), thalamus, and to the prefrontal cortex are thought to comprise the neural basis of our abilities to interpret social behaviors. Larger amygdala volumes were previously reported in pediatric GAD; however, the brain regions involved in social intelligence were not examined in this pilot study. Methods: Magnetic resonance imaging (MRI) was used to measure the STG, thalamus, and prefrontal volumes in 13 medically healthy child and adolescent subjects with generalized anxiety disorder (GAD) and 98 comparison subjects, who were at low familial risk for mood and psychotic disorders. Groups were similar in age, gender, height, weight, handedness, socioeconomic status, and full-scale IQ. Results: The total, white matter, and gray matter STG volumes were significantly larger in GAD subjects compared with control subjects. Thalamus and prefrontal lobe volumes did not differ between groups. Findings of significant side-by-diagnosis interactions for STG and STG white matter volumes suggest that there is a more pronounced right ⬎ left asymmetry in total and STG white matter volumes in pediatric GAD subjects compared with control subjects. A significant correlation between the STG white matter percent asymmetry index with the child report of the Screen for Child Anxiety Related Emotional Disorders Scale was seen. Conclusions: These data agree with previous work implicating posterior right-hemispheric regions in anxiety dis-

From the University of Pittsburgh Medical Center (MDDB, MSK, RED, DAA, BB, NDR), Developmental Traumatology Program (Developmental Family Health Clinic and Neuroimaging Laboratory), Western Psychiatric Institute and Clinic (MDDB, JH, GM), University of Pittsburgh Medical Center; and Department of Statistics (SI, HS), University of Pittsburgh, Pittsburgh, Pennsylvania. Address reprint requests to Michael D. De Bellis, M.D., M.P.H., University of Pittsburgh Medical Center, Developmental Traumatology Program (Developmental Family Health Clinic and Neuroimaging Laboratory), Western Psychiatric Institute and Clinic, Room 392, 3811 O’ Hara Street, Pittsburgh PA 15213. Received September 26, 2001; revised December 7, 2001; accepted December 13, 2001.

© 2002 Society of Biological Psychiatry

orders and may suggest developmental alterations in pediatric GAD. Biol Psychiatry 2002;51:553–562 © 2002 Society of Biological Psychiatry Key Words: Generalized anxiety disorder, thalamus, superior temporal gyrus, neurodevelopment, pediatric anxiety disorders

Introduction

T

he essential symptoms of generalized anxiety disorder (GAD) are intrusive worry about everyday life circumstances and social competence, and associated autonomic hyperarousal. The amygdala, a brain region involved in fear and fear-related behaviors in animals, and its projections to the superior temporal gyrus (STG), thalamus, and to the prefrontal cortex are thought to comprise the neural basis of our abilities to interpret other’s behavior in terms of mental states (e.g., thoughts, intentions, desires, beliefs) or social intelligence (Brothers 1990). The STG and amygdala are involved in processing social information (Baron-Cohen et al 1999). In primate studies, cells that are involved in identifying facial expressions are located in the STG (Desimone 1991; Hasselmo et al 1989). In support of this idea, results from a recently published functional magnetic resonance imaging (MRI) study demonstrated that the amygdala, STG, and prefrontal cortex were activated during the performance of a social intelligence task in healthy volunteers (BaronCohen et al 1999). In studies of experimental conditioning, the STG is thought to be involved in higher cognitive processing of the fear experience and modulation of amygdala activity (Quirk et al 1997). Recently, significantly larger right and total amygdala volumes were reported in a small sample of pediatric subjects with GAD compared with healthy control subjects, whereas intracranial, cerebral, cerebral gray and white matter, temporal lobe, hippocampal, and basal ganglia volumes and measures of the midsagittal area of the corpus callosum did not differ between groups (De Bellis et al 2000); however, the brain regions involved in 0006-3223/02/$22.00 PII S0006-3223(01)01375-0

554

M.D. De Bellis et al

BIOL PSYCHIATRY 2002;51:553–562

Table 1. Demographic Characteristics of Children and Adolescents with GAD and Healthy Control Subjects Characteristic n Age (years) (range in years) Height (cm) Weight (kg) Race White/African American/biracial SES (range) Sex (female/male) Verbal IQ (range) Performance IQ (range) Full-scale IQ (range)

GAD

Healthy Control Subjects

Statistic

p

13 12.5 ⫾ 2.5 (8.0 –15.8) 60.60 ⫾ 4.69 (54.9 –70.90) 54.2 ⫾ 20.5 (34.5–103.6)

98 12.0 ⫾ 2.3 (7.6 –16.25) 60.18 ⫾ 6.16 (42–74.5) 48.4 ⫾ 17.5 (16.3–105.7)

— t109 ⫽ ⫺.74 t109 ⫽ ⫺.24 t109 ⫽ ⫺1.11

— .461 .813 .269

8/4/1 41.3 ⫾ 15.4 (17– 64) 5/8 119.9 ⫾ 15.2 (100 –145) 125.9 ⫾ 15.8 (100 –155) 125.1 ⫾ 16.0 (100 –148)

75/9/14 41.4 ⫾ 9.8 (18 – 64) 48/50 114.4 ⫾ 13.5 (87–147) 119.1 ⫾ 17.4 (79 –152) 118.2 ⫾ 15.3 (89 –153)

FET t109 ⫽ .03 ␹2 ⫽ .509 t109 ⫽ ⫺1.36 t109 ⫽ ⫺1.34 t109 ⫽ ⫺1.50

.07 .976 .476 .176 .182 .136

GAD, generalized anxiety disorder; PTSD, posttraumatic stress disorder; FET, Fisher’s exact test; SES, socioeconomic status.

social intelligence (STG, thalamus, and prefrontal cortex volumes) were not examined in this pilot study. Interestingly, a finding of relatively larger STG, and particularly right-sided STG gray matter volumes, was recently reported in pediatric patients with posttraumatic stress disorder (PTSD) that was secondary to child maltreatment, a social trauma, compared with nonabused control subjects (De Bellis et al, this volume). Generalized anxiety disorder, posttraumatic stress disorder, and obsessive-compulsive disorder (OCD) are similar in that each is manifested by symptoms of intrusive thoughts. Pediatric OCD is associated with larger thalamus volumes (Gilbert et al 2000) and smaller striatum structures (caudate and putamen) (Rosenberg et al 1997b) and a selective deficit in frontostriatal function (Rosenberg et al 1997a). In this study, we examined STG, thalamus, and prefrontal cortex volumes of 13 children and adolescents with GAD and 98 sociodemographically similar comparison subjects who were at low familial risk for mood and psychotic disorders. Both groups had no personal history of trauma or maltreatment. We hypothesized that STG and thalamus volumes would be greater in GAD subjects than for control subjects. Given the data showing greater activity in the right hemisphere of anxious persons (for review, see Davidson et al 1999), we also sought to explore morphologic right/left asymmetry measures in these subjects.

Methods and Materials Subjects Children and adolescents with DSM-IV GAD (n ⫽ 13) and healthy comparison subjects (n ⫽ 98) were recruited. Groups were similar in age, gender, height, and handedness, weight, socioeconomic status, and full-scale IQ. This study is part of ongoing investigations on the functional neurobiology of childhood mood and anxiety disorders and the neurobiology of childhood maltreatment. Measures of intracranial, cerebral, cerebral gray, cerebral white matter, temporal lobe, amygdala,

hippocampal, caudate and putamen volumes, and corpus callosum area were previously reported in 12 of the GAD subjects and 24 of the control subjects (De Bellis et al 2000). Subjects underwent the Hollingshead Four Factor Index of Socioeconomic Status (SES; Hollingshead 1975) for an assessment of SES; the vocabulary, digit span, block design, and object assembly subsets of the Wechsler Intelligence Scale for Children (WISC-R) for an estimate of IQ (Wechsler 1974); and the 12 handedness items from the Revised Physical and Neurologic Examination for Subtle Signs (PANESS) Inventory (Denckla 1985) where 8 of 12 items defined all subjects as right-handed. The demographic characteristics of the groups are found in Table 1. Anxious children were recruited from the Child and Adolescent Anxiety and Depression Program of the Western Psychiatric Institute and Clinic, University of Pittsburgh. Healthy control children and adolescents were recruited (n ⫽ 122) for ongoing pediatric neuroimaging investigations by the Developmental Traumatology Program at Western Psychiatric Institute and Clinic. The MRI scans of control children and adolescents (n ⫽ 98), who were at low familial risk for mood and psychotic disorders, who had a full-scale IQ ⬎ 80, and who were sociodemographically similar to the GAD subjects, were selected for comparisons purposes. Control subjects were required never to have had any lifetime psychopathology. Control children were also required to have no first-degree relatives with a lifetime episode of any mood or psychotic disorder; no second-degree relatives with a lifetime history of childhood-onset, recurrent, psychotic, or bipolar depression, schizoaffective, or schizophrenic disorder; and no more than 20% of their second-degree relatives could have a lifetime single episode of major depression. Anxious and control children were evaluated by trained research clinicians blind to the subjects’ clinical status under the supervision of child psychiatrists (DAA, BB, or MDDB), using a modified version of the Schedule for Affective Disorders and Schizophrenia for School-Age, Present Episode (K-SADS-P; Chambers et al 1985) and Lifetime Version (K-SADS-E; Orvaschel and Puig-Antich 1987) interview with both child and parent(s) as informants. Questions concerning traumatic events and posttraumatic stress disorder (PTSD) symptoms over the subject’s lifetime were incorporated into an expanded assessment

Superior Temporal Gyrus Volumes in Pediatric GAD

of PTSD completed as part of the K-SADS. Additional questions involving the types of interpersonal and noninterpersonal traumas and the nature and circumstances of such traumatic experiences are described (Kaufman et al 1997). Comorbidity in the GAD group included the following: depressive disorder NOS (n ⫽ 3), major depression (n ⫽ 1), panic disorder (n ⫽ 1), and social phobia (n ⫽ 1). The Screen for Child Anxiety Related Emotional Disorders Scale (SCARED; Birmaher et al 1999), a 41-item parent and child self-report instrument, was given to GAD subjects. The SCARED consists of five factors that parallel the DSM-IV classification of anxiety disorders: somatic/panic, generalized anxiety, separation anxiety, social phobia, and school phobia (Birmaher et al 1997) and was used as a continuous measure of childhood anxiety symptoms. Total score for child report was 28.17 ⫾ 13.93; for the parent report of the child, total score was 34.91 ⫾ 15.78. These are above the clinical cutoff of 25 (Birmaher et al 1999). One child with GAD and two parents of GAD children did not complete the SCARED. All GAD subjects were psychotropic-naı¨ve except one who had a history of brief treatment with antidepressants and a stimulant before the MRI scan. Exclusionary criteria were as follows: 1) the use or presence of medication with central nervous system or hypothalamicpituitary effects within the previous 2 weeks; 2) presence of a significant medical or neurologic illness; 3) obesity (weight greater than 150% of ideal body weight) or growth failure (height or weight under third percentile); 4) full-scale IQ lower than 80; 5) anorexia nervosa, autism, or schizophrenia by DSM-IV criteria; 6) inordinate fear of intravenous needles; 7) GAD chronologically secondary to conduct disorder; 8) specific learning disabilities; 9) a current diagnosis of PTSD or severe maltreatment history (with Child Protective Services involvement) in the GAD group; 10) positive trauma or maltreatment history in subjects in the healthy control group; and 11) adolescent-onset alcohol or substance abuse or dependence; in addition, 12) subjects were screened for any contraindication for MRI scans (floating metallic bodies, severe claustrophobia). This protocol was approved by the University of Pittsburgh Institutional Review Board. Parent(s) or legal guardian(s) gave written informed consent. Children under age 14 years assented before participating in this protocol. Adolescents, 14 years of age and older, gave written informed consent along with the written informed consent of their parent or legal guardian. No subject consented to participate independently of a parent or legal guardian. Subjects received monetary compensation for participation.

MRI Acquisition Magnetic resonance imaging was performed using a GE 1.5 Tesla Unit (Signa System, General Electric Medical Systems, Milwaukee, WI) running version 5.4 software located at the UPMC MR Research Center. A three-dimensional spoiled gradient recalled acquisition in the steady-state pulse sequence was used to obtain 124 contiguous images with slice thickness of 1.5 mm in the coronal plane. (using TE ⫽ 5 msec, TR ⫽ 25 msec, flip angle ⫽ 40°, acquisition matrix ⫽ 256 ⫻ 192, NEX ⫽ 1,

BIOL PSYCHIATRY 2002;51:553–562

555

FOV ⫽ 24 cm). Coronal sections were obtained perpendicular to the anterior-commissure–posterior-commissure line to provide a more reproducible guide for image orientation. No sedation was used. Further details concerning desensitization to scanner procedure, head placement, imagine transfer, and structural analysis of cerebral and prefrontal lobe volumes were previously described by our group (De Bellis et al 1999, 2000), including standardization of head position and quantification of gray and white matter (De Bellis et al 2001b) and are briefly described here. The imaging data were transferred from the MRI unit to a computer workstation (Power Macintosh, Apple Computer) and analyzed using IMAGE software (version 1.62) developed at the National Institutes of Health (Rasband 1996) that provides valid and reliable volume measurements of specific structures using a manually operated (hand-tracing) approach. All measurements were made by trained and reliable raters who were blind to subject information. Intraclass correlation of interrater and intrarater reliability for independent designation of individual brain regions were obtained on segmented images from measurements of 20 subject scans for each brain structure reported. Intraclass correlation of interrater reliability ranged from 0.98 to 0.99 for intracranial, cerebral, cerebral gray matter, cerebral white matter, prefrontal lobe volume, prefrontal lobe gray matter, and prefrontal lobe white matter (JH and AMB) and for right and left STG volumes, anterior and posterior STG volumes, and right and left STG volumes gray and white matter volumes (JH and KF). Intrarater reliabilities were 0.99 for all measures including the thalamus. Brain reliability measures were examined within a week and across time (3– 6 months). Measures of the STG and thalamus were based on neuroanatomic boundaries determined by standard neuroanatomic atlases (Roberts and Hanaway 1971; Yuh et al 1994) and based on previously published child and adolescent neuroimaging studies using manually traced methods for the coronal plane (Giedd et al 1996; Gilbert et al 2000; Jacobsen et al 1998). The STG was defined as the gyrus boundary in each of the coronal sections. The anterior boundary of the STG was identified as the first slice showing the white matter tract (temporal stem) connecting the temporal lobe with the base of the brain. The appearance of the fibers of the crux of the fornix served as the posterior boundary (Shenton et al 1992). Every coronal slice that included the STG was manually traced, and the volume was computed by summing up successive areas and multiplying by slice thickness. The STG was further divided into anterior and posterior boundaries using the slice showing the most anterior mammillary bodies to define this boundary. These posterior STG measures included the planum temporale and the majority of (but not the most posterior coronal slices) of Heschl gyrus as described (De Bellis et al, this volume). The boundaries of the thalamus were traced as previously described (Gilbert et al 2000). Briefly, the anterior boundary of the thalamus was defined by the mammillary bodies and the interventricular foramen. The internal capsule was considered the lateral boundary, the third ventricle, the medial boundary, and the hypothalamus, the inferior boundary. The superior boundary was defined by the lateral ventricle and the posterior boundary was defined by the crus fornix. Every coronal slice that included the thalamus was manually traced, and total volume was com-

556

M.D. De Bellis et al

BIOL PSYCHIATRY 2002;51:553–562

Table 2. Prefrontal Cortex, Thalamus, and Superior Temporal Gyrus Volumes of Children and Adolescents with GAD and Healthy Control Subjects Measure (cm3)

GAD (mean ⫾ SD)

Control (mean ⫾ SD)

TCV CGV CWV Frontal volume

1281.5 ⫾ 94.8 807.3 ⫾ 80.2 474.2 ⫾ 45.6 176.29 ⫾ 18.72

1270 ⫾ 155 810 ⫾ 108 459.8 ⫾ 80.1 181.38 ⫾ 29.74

t109 ⫽ ⫺.27 p ⫽ .79 t109 ⫽ .08 p ⫽ .94 t109 ⫽ ⫺.63 p ⫽ .53 t109 ⫽ .60 p ⫽ .55

Frontal gray

116.43 ⫾ 14.81

120.98 ⫾ 19.61

t109 ⫽ ⫺1.19

Frontal white

59.85 ⫾ 7.90

60.40 ⫾ 15.42

t109 ⫽ ⫺.64

p ⫽ .52

Frontal CSF

3.43 ⫾ 1.13

3.82 ⫾ 1.89

t109 ⫽ ⫺.71

p ⫽ .48

Total thalamus

7.50 ⫾ 1.04

7.65 ⫾ 1.82

t109 ⫽ .29

p ⫽ .77

Right thalamus

3.83 ⫾ .56

3.92 ⫾ .923

t109 ⫽ .33

p ⫽ .75

Left thalamus

3.67 ⫾ .55

3.74 ⫾ .96

t109 ⫽ .23

p ⫽ .82

191.17 ⫾ 19.14

190.16 ⫾ 21.88

t109 ⫽ ⫺.16

p ⫽ .87

Right temporal lobe

100.22 ⫾ 11.24

97.74 ⫾ 11.44

t109 ⫽ ⫺.74

p ⫽ .46

Left temporal lobe

90.95 ⫾ 9.33

92.41 ⫾ 11.60

t109 ⫽ .43

51.59 ⫾ 6.91

45.55 ⫾ 5.21

t109 ⫽ ⫺3.78

p ⫽ .0003

Total right STGc

27.06 ⫾ 3.94

23.36 ⫾ 2.68

t109 ⫽ ⫺4.42

p ⬍ .0001

Total right STG grayb

19.01 ⫾ 3.06

16.98 ⫾ 2.41

t109 ⫽ ⫺2.77

p ⫽ .0067

Total right STG whitec

7.96 ⫾ 1.60

6.38 ⫾ 1.53

t109 ⫽ ⫺3.48

p ⫽ .001

Total left STGb

24.53 ⫾ 3.37

22.19 ⫾ 2.75

t109 ⫽ ⫺2.81

p ⫽ .006

Total left STG grayb

19.37 ⫾ 2.92

17.79 ⫾ 2.55

t109 ⫽ ⫺2.07

p ⫽ .041

Total left STG white

5.05 ⫾ 1.45

4.40 ⫾ 1.04

t109 ⫽ ⫺2.02

p ⫽ .046

Anterior right STGc

11.45 ⫾ 2.33

9.74 ⫾ 1.46

t109 ⫽ ⫺3.66

p ⫽ .0004

Anterior right STG grayc

9.23 ⫾ 1.72

8.04 ⫾ 1.24

t109 ⫽ ⫺3.10

p ⫽ .0025

Anterior right STG whiteb

2.221 ⫾ 0.736

1.702 ⫾ 0.623

t109 ⫽ ⫺2.76

p ⫽ .007

Posterior right STGc

15.53 ⫾ 2.55

13.62 ⫾ 1.71

t109 ⫽ ⫺3.55

p ⫽ .001

Posterior right STG grayb

9.79 ⫾ 1.78

8.94 ⫾ 1.53

t109 ⫽ ⫺1.84

p ⫽ .068

Posterior right STG whitec

5.74 ⫾ 1.19

4.67 ⫾ 1.03

t109 ⫽ ⫺3.44

p ⫽ .001

10.13 ⫾ 2.17

9.19 ⫾ 1.57

t109 ⫽ ⫺1.93

p ⫽ .056

Anterior left STG gray

9.13 ⫾ 1.88

8.39 ⫾ 1.50

t109 ⫽ ⫺1.63

p ⫽ .107

Anterior left STG white

.998 ⫾ 0.496

.799 ⫾ 0.325

t109 ⫽ ⫺1.93

p ⫽ .056

Posterior left STGb

14.3 ⫾ 2.18

13.0 ⫾ 1.66

t109 ⫽ ⫺2.53

p ⫽ .013

Temporal lobes

Total STGc

Anterior left STGb

t109 value p value

p ⫽ .24

p ⫽ .66

Covariate, t1,108 value, p valued

Group: t ⫽ ⫺1.32, p ⫽ .19 CV: t ⫽ 13.47, p ⬍ .0001 Group: t ⫽ ⫺1.20, p ⫽ .23 CGV: t ⫽ 13.25, p ⬍ .0001 Group: t ⫽ ⫺.64, p ⫽ .52 CWV: t ⫽ 8.04, p ⬍ .0001 Group: t ⫽ ⫺.70, p ⫽ .49 CV: t ⫽ ⫺.88, p ⫽ .3823 Group: t ⫽ .29, p ⫽ .77 CGV: t ⫽ 6.44, p ⬍ .0001 Group: t ⫽ .33, p ⫽ .74 CGV: t ⫽ 5.93, p ⬍ .0001 Group: t ⫽ .22, p ⫽ .83 CGV: t ⫽ 6.33, p ⬍ .0001 Group: t ⫽ ⫺.08, p ⫽ .94 CV: t ⫽ 15.13, p ⬍ .0001 Group: t ⫽ ⫺1.01, p ⫽ .31 CV: t ⫽ 13.32, p ⬍ .0001 Group: t ⫽ .85, p ⫽ .40 CV: t ⫽ 12.71, p ⬍ .0001 Group: t ⫽ 4.18, p ⬍ .0001 TCV: t ⫽ 6.00, p ⬍ .0001 Group: t ⫽ 4.97, p ⬍ .0001 TCV: t ⫽ 6.29, p ⬍ .0001 Group: t ⫽ 3.28, p ⫽ .0014 CGV: t ⫽ 6.36, p ⬍ .0001 Group: t ⫽ 4.39, p ⬍ .0001 CWV: t ⫽ 11.08, p ⬍ .0001 Group: t ⫽ 2.98, p ⫽ .0036 TCV: t ⫽ 5.09, p ⬍ .0001 Group: t ⫽ 2.50, p ⫽ .0140 CGV: t ⫽ 6.68, p ⬍ .0001 Group: t ⫽ 1.95, p ⫽ .0541 CWV: t ⫽ 5.98, p ⬍ .0001 Group: t ⫽ 3.85, p ⬍ .0001 TCV: t ⫽ 4.55, p ⬍ .001 Group: t ⫽ 3.53, p ⫽ .0006 CGV: t ⫽ 5.52, p ⬍ .0001 Group: t ⫽ 2.92, p ⫽ .0043 CWV: t ⫽ 7.38, p ⬍ .0001 Group: t ⫽ 3.84, p ⫽ .0002 TCV: t ⫽ 5.37, p ⬍ .0001 Group: t ⫽ 2.09, p ⫽ .0394 CGV: t ⫽ 5.15, p ⬍ .0001 Group: t ⫽ 4.27, p ⬍ .0001 CWV: t ⫽ 10.75, p ⬍ .0001 Group: t ⫽ 1.95, p ⫽ .0532 TCV: t ⫽ 3.91, p ⫽ .0002 Group: t ⫽ 1.84, p ⫽ .0682 CGV: t ⫽ 5.09, p ⬍ .0001 Group: t ⫽ 1.85, p ⫽ .0674 CWV: t ⫽ 5.73, p ⬍ .0001 Group: t ⫽ 2.62, p ⫽ .0102 TCV: t ⫽ 4.29, p ⬍ .0001

Superior Temporal Gyrus Volumes in Pediatric GAD

BIOL PSYCHIATRY 2002;51:553–562

557

Table 2. Continued GAD (mean ⫾ SD)

Control (mean ⫾ SD)

Posterior left STG gray

10.24 ⫾ 1.74

9.40 ⫾ 1.42

t109 ⫽ ⫺1.95

p ⫽ .054

Posterior left STG white

4.05 ⫾ 1.1

3.604 ⫾ 0.791

t109 ⫽ ⫺1.83

p ⫽ .069

38.39 ⫾ 5.75

34.77 ⫾ 4.81

t109 ⫽ ⫺2.49

p ⫽ .014

Total anterior STG grayb

18.36 ⫾ 3.44

16.43 ⫾ 2.61

t109 ⫽ ⫺2.41

p ⫽ .017

Total posterior STG grayb

20.03 ⫾ 3.35

18.34 ⫾ 2.82

t109 ⫽ 1.98

13.01 ⫾ 2.76

10.78 ⫾ 2.24

t109 ⫽ ⫺3.29

p ⫽ .001

Total anterior STG whiteb

3.22 ⫾ 1.07

2.501 ⫾ 0.829

t109 ⫽ ⫺2.83

p ⫽ .006

Total posterior STG whitec

9.79 ⫾ 2.15

8.28 ⫾ 1.59

t109 ⫽ ⫺3.08

p ⫽ .003

Measure (cm3) b

Total STG grayb

Total STG whitec

t109 value p value

p ⫽ .050

Covariate, t1,108 value, p valued Group: t ⫽ 2.27, p ⫽ .0249 CGV: t ⫽ 5.91, p ⬍ .0001 Group: t ⫽ 1.72, p ⫽ .0875 CWV: t ⫽ 5.27, p ⬍ .0001 Group: t ⫽ 3.02, p ⫽ .0031 CGV: t ⫽ 6.85, p ⬍ .0001 Group: t ⫽ 2.77 p ⫽ .0065 CGV: t ⫽ 5.62, p ⬍ .0001 Group: t ⫽ 2.31, p ⫽ .0227 CGV: t ⫽ 5.86, p ⬍ .0001 Group: t ⫽ 4.08, p ⬍ .0001 CWV: t ⫽ 10.93, p ⬍ .0001 Group: t ⫽ 3.10, p ⫽ .0025 CWV: t ⫽ 8.33, p ⬍ .0001 Group: t ⫽ 3.59, p ⫽ .0005 CWV: t ⫽ 9.60, p ⬍ .0001

GAD, generalized anxiety disorder; CGV, cerebral gray volume; CSF, cerebrospinal fluid; CV, cerebral volume; CWV, cerebral white volume; STG, superior temporal gyrus; TCV, total cerebral volume. a p ⬍ .05. b Statistically significant for both group and brain structure (p ⬍.05). c Bonferroni criterion for significance, p ⬍ .001. d Regression analysis.

puted by summing up successive areas and multiplying by slice thickness.

Statistical Methods Histograms were used to check for outliers and normality of the data. Correlation between variables was checked against a matrix of scatterplots. Because of the large number of variables, a principal components analysis was performed to reduce the dimensionality of the MRI data. Formal hypothesis testing regarding differences between groups and STG, thalamus, and prefrontal volumes consisted of a hierarchy of analyses starting with t tests and then a more complicated regression analysis. Regression analysis was used to look at group differences for all dependent variables with appropriate brain structure as covariates. For example, when examining group differences between total STG volumes, cerebral volume was used as a covariate. When examining group differences between total STG gray matter, cerebral gray matter was used as a covariate. Lastly, when examining group differences between total STG white matter, cerebral white matter was used as a covariate. Percent asymmetry in right versus left STG measures were defined as (R ⫺ L) / (R ⫹ L) ⫻ 100. In testing for group differences in the normal right ⬎ left STG asymmetry, right and left STG volumes were analyzed by two-way repeated-measures analyses of covariance with group as the between-subjects factor, side (right and left) as the repeated factor, and appropriate brain structure as the covariate. Next, correlations with clinical variables were calculated for STG structural means differing significantly between the groups. A more involved regression analysis examined age, gender, group, brain structure, and interactions on the same STG-dependent variables and principal component scores. A final regression was performed to determine the impact of

clinical variables on the variables for which group differences existed. All significance testing was two-tailed with alpha ⫽ 0.05. All data are presented as mean ⫾ SD (SD) unless otherwise specified. Because of the large number of structural comparisons, we set alpha ⫽ 0.01 for exploratory regression involving clinical correlations.

Results Histograms of the STG-, thalamus-, and prefrontal-dependent variables showed that all variables were reasonably symmetric approximating normality. The matrix of scatterplots showed that the variables were linearly related. The principal component analysis showed that approximately 88% of the variability of the data could be explained by using three principal components. Principal component 1 can be interpreted as a general average of all the STG measurements, principal component 2 can be interpreted as a contrast between the gray matter and white matter STG measurements, and principal component 3 can be interpreted as a contrast between anterior and posterior STG measurements. These three principal components were used later in regression analysis. The t tests showed that total STG, total STG gray matter, and total STG white matter volumes were larger in GAD subjects than in control subjects. Guided statistical analyses were done to further clarify the differences between groups (e.g., anterior vs. posterior STG, right vs. left STG, STG gray vs. STG white matter volumes). Note that total STG, total STG right (anterior

558

M.D. De Bellis et al

BIOL PSYCHIATRY 2002;51:553–562

and posterior), and posterior right STG volumes, anterior right STG gray matter volume, and total STG white and total STG posterior white, total right STG white matter, and right STG posterior white matter volumes met the more conservative Bonferroni criterion for significance (p ⬍ .001). When covarying for group and appropriate brain structure, regression analysis showed that group and brain structure remained significant in all STG measurements. Because IQ may predict STG volumes, we preformed Spearman’s rho correlations between IQ and STG structures. Verbal (rs ⫽ 37, df ⫽ 109, p ⬍ .0001), performance (rs ⫽ 21, df ⫽ 109, p ⫽ .01) and full-scale IQ (rs ⫽ 32, df ⫽ 109, p ⬍ .001) did significantly predict STG volumes; however, when covarying for group, appropriate brain structure, and verbal IQ, regression analysis showed that group remained significant in all STG measurements (e.g., STG volume: F ⫽ 17.26, df ⫽ 107, p ⬍ .0001; right STG: F ⫽ 25.64, df ⫽ 107, p ⬍ .0001; STG gray matter: F ⫽ 8.47, df ⫽ 107, p ⫽ .004; and STG white matter: F ⫽ 16.36, df ⫽ 107, p ⬍ .0001). Thalamic and prefrontal volumes did not differ between groups; see Table 2 and Figure 1. There were significant side-by-diagnosis interactions for STG and STG white and posterior STG white but not gray matter volumes. These findings suggest that there is a more pronounced right ⬎ left asymmetry in total (anterior and posterior) STG volumes, STG white and posterior STG white matter volumes. See Table 3 and

Figure 1. Total superior temporal gyrus (STG), total STG gray matter, and total STG white matter volumes and means (cm3) of healthy comparison subjects (squares) and children and adolescents with generalized anxiety disorder (GAD; diamonds ⫽ GAD without comorbid mood disorder, circles ⫽ GAD with comorbid mood disorder). The GAD subjects had significantly larger total STG volumes (t109 ⫽ ⫺3.78, p ⫽ .0003), STG gray matter volumes (t109 ⫽ ⫺2.49, p ⫽ .014), and STG white matter volumes (t109 ⫽ ⫺3.29, p ⫽ .001) than control subjects.

Figure 2. This is further described by a significant increase (t109 ⫽ ⫺2.23, p ⬍ .03) seen in the STG percent asymmetry index in GAD (4.86 ⫾ 4.67) versus control subjects (2.61 ⫾ 3.24) and the suggestion of a significant increase in the STG white percent asymmetry index (t109 ⫽ ⫺1.51, p ⫽ .1) in GAD (23.01 ⫾ 10.1) versus control subjects (18.11 ⫾ 11.18). No significant differences were seen in STG gray matter percent asymmetry index (t109 ⫽ ⫺1.19, p ⫽ .24) in GAD versus control subjects. The STG white matter percent asymmetry index significantly correlated with the child report on the SCARED (r ⫽ 0.68, df ⫽ 10, p ⫽ .01), but not the parent report on the SCARED (r ⫽ 0.03, df⫽ 9, p ⫽ .97). No other significant correlations were seen between STG volumes and clinical variables (SCARED scores) in the GAD group; see Figure 3.

Discussion Children and adolescents with GAD were found to have significantly larger STG total and gray and white matter volumes than sociodemographically similar control subjects. Prefrontal cortex and thalamus volumes did not differ between groups. Furthermore, findings of significant side-by-diagnosis interactions for STG and STG white matter volumes suggest that there is a more pronounced right ⬎ left asymmetry in total and white matter STG volumes in GAD subjects compared with control subjects. The STG white matter percent asymmetry index significantly correlated with self-reports of anxiety symptoms. We previously reported that right and total amygdala volumes were significantly larger in 12 of these GAD subjects compared with control subjects, whereas intracranial, cerebral, cerebral gray and white matter, temporal lobe, hippocampal, and basal ganglia volumes and measures of the midsagittal area of the corpus callosum did not differ between groups (De Bellis et al 1999). To our knowledge, this is the first study to report a structural difference in the STG in pediatric GAD. These data may suggest that there are developmental alterations in pediatric GAD, manifested by greater anatomical right ⬎ left asymmetry in the STG and larger right amygdala volumes with no significant anatomic differences in MRI measures of prefrontal, basal ganglion, and thalamus volumes. These data are similar to a recent finding in pediatric maltreatment-related PTSD in which significantly larger right, left and total STG and STG gray volumes and findings of significant side-by-diagnosis interactions for STG volumes suggest that there is a more pronounced right ⬎ left asymmetry in total and posterior STG volumes, but no differences in prefrontal, hippocampal, or basal ganglion structures (De Bellis et al, this volume);

Superior Temporal Gyrus Volumes in Pediatric GAD

BIOL PSYCHIATRY 2002;51:553–562

559

Table 3. Repeated Measures ANCOVA for Group, Side, and Group-by-Side Interactions for Superior Temporal Gyrus (STG) Structures of 13 Children and Adolescents with GAD and 98 Healthy Control Subjects Group

Group ⫻ Side

Side

Measure (cm3)

F1,108

p

F1,108

p

F1,108

p

STG STG gray STG white Anterior STG Anterior STG gray Anterior STG white Posterior STG Posterior STG gray Posterior STG white

18.89 9.14 16.64 9.97 7.69 9.61 12.47 5.34 12.86

⬍.0001 .003 ⬍.001 ⬍.004 ⬍.007 ⬍.003 .0006 ⬍.03 .0005

.17 .04 2.09 .02 .39 .23 .72 .11 3.44

.67 .84 .15 .90 .54 .63 .40 .74 ⬍.07

8.04 1.35 5.17 5.37 2.78 3.30 4.68 .0004 5.02

.006 .25 .02 .02 ⬍.10 .07 ⬍.04 .98 ⬍.03

Comment GAD GAD GAD GAD GAD GAD GAD GAD GAD

⬎ ⬎ ⬎ ⬎ ⬎ ⬎ ⬎ ⬎ ⬎

control, control, control, control, control, control, control, control, control,

right right right right right right right right right

and and and and and and and and and

left. left. left. left. left. left. left. left. left.

Effect larger on the right. Effect larger on the right. Effect larger on the right.

Effect larger on the right. Effect larger on the right.

GAD, generalized anxiety disorder; ANCOVA, analyses of covariance.

however, the amygdala volumes of pediatric subjects with maltreatment-related PTSD did not differ from control subjects in cross-sectional (De Bellis et al 1999) or in longitudinal studies (De Bellis et al 2001a). Taken together, these data suggest that there are developmental alterations in the STG in pediatric anxiety disorders that may not necessarily be related to having a trauma history; however, the nature of the larger STG findings differed between pediatric subjects with PTSD (STG gray matter asymmetry) and GAD (STG white matter asymmetry). Furthermore, subjects with GAD had relatively larger STG volumes (GAD: 56.82 ⫾ 13.17 cm3 vs. control: 44.85 ⫾ 9.09 cm3, means adjusted for cerebral volumes) than maltreated subjects with PTSD (PTSD 50.58 ⫾ 10.17 cm3 vs. control: 46.32 ⫾ 6.79 cm3, means adjusted for cerebral volumes; De Bellis et al, this volume). These results also differ from findings that pediatric OCD is associated with larger thalamus volumes (Gilbert et al 2000) and smaller

Figure 2. Least squares means plots of significant side-bydiagnosis (subjects with generalized anxiety disorder [GAD] ⫽ dashed line, control subjects ⫽ solid line) interactions for (A) right and left superior temporal gyrus (STG) volumes (F1,108 ⫽ 8.04, p ⫽ .006) and (C) right and left STG white matter volumes (F1,108 ⫽ 5.17, p ⫽ .02) but not (B) gray matter volumes (F1,108 ⫽ 1.35, p ⫽ .25). These findings suggest that there is a more pronounced right ⬎ left asymmetry in STG volumes in children and adolescents with GAD compared with healthy comparison subjects.

striatum structures (caudate and putamen; Rosenberg et al 1997b) and a selective deficit in frontostriatal function (Rosenberg et al 1997a) in medication-naı¨ve subjects. To date, studies of STG measures have not been reported in pediatric OCD. Thus, each pediatric anxiety disorder subtype may have its own unique neurobiology. Our findings suggested a more pronounced right ⬎ left asymmetry in total and white matter STG volumes. Anatomic findings do not necessarily imply functional changes, yet these asymmetry findings are interesting in light of the results of electrophysiologic and imaging studies in humans. There is a substantial literature indicating the involvement of the right cerebral hemisphere in identifying facial expressions (Etcoff 1986). Several studies have demonstrated right-sided prefrontal activation in

Figure 3. Significant correlation between the superior temporal gyrus (STG) white matter percent asymmetry index with the child report of the Screen for Child Anxiety Related Emotional Disorders Scale (r ⫽ .68, df ⫽ 10, p ⫽ .01).

560

M.D. De Bellis et al

BIOL PSYCHIATRY 2002;51:553–562

temperamentally fearful monkeys (Kalin et al 1998) and in behaviorally inhibited children (for review, see Kagan and Snidman 1999). Behaviorally inhibited children are predisposed to develop childhood anxiety disorders (Biederman et al 1993). Several electroencephalography investigations have shown greater electrical activity in the right hemispheres of anxious and anxious– depressed adults, and especially greater posterior right-hemispheric activity in persons with a high dimension of anxious arousal (Davidson et al 1999; Heller 1993; Heller et al 2000; Keller et al 2000). Higher rates of glucose metabolism were seen in the anxious individuals in the right hemisphere during an anxiety producing task (Wu et al 1991). Results from recent functional MRI investigations showed right-hemisphere dominant activation in the amygdala– periamygaloid cortex during both conditioned fear acquisition and extinction in healthy subjects (LaBar et al 1998). Furthermore, results from a recent positron emission tomography (PET) study showed that healthy subjects with high trait anxiety had greater right–left ratios of cerebral metabolism than low trait anxiety subjects, particularly during the second PET session (Stapleton et al 1997). In this study, a significant correlation was seen between the STG white matter percent asymmetry index and the child report on the SCARED. Greater desynchronization of alpha frequencies in the right frontal region may, in part, reflect activity of cholinergic projections to the basal nucleus of Meynert originating from greater activity of the right STG and amygdala (Lloyd and Kling 1991). Perhaps anxious children without trauma histories have an inherent propensity for larger right amygdala volumes and “connectivity” to larger STG volumes, creating some predispositional trait such as increased sensitivity to social cues with anxious arousal, whereas maltreated children may be “conditioned” to be more fearful of social cues. Anxiety disorders are often comorbid with depression disorders. Depressed adolescents and adults without comorbid anxiety disorders demonstrate reduced left frontal and right parietal electroencephalographic activation, whereas depressed adolescents and adults with comorbid anxiety disorders demonstrate evidence of greater activation of right and parietal sites (Keller et al 2000; Kentgen et al 2000); however, adults but not adolescents with comorbid anxiety disorders exhibited a reduced right ear–left hemisphere advantage for fused words in a dichotic listening task, suggesting developmental differences (Pine et al 2000). Furthermore, depressed adults with greater left- than right-hemispheric processing of dichotic stimuli show more favorable antidepressant treatment responses than depressed adults with overall greater right-hemispheric activation (Bruder et al 2001). The presence of a comorbid mood disorder did not seem to

influence the reported STG findings; however, behaviorally inhibited children are predisposed to depression during young adulthood (Caspi et al 1996), thus prospective studies of anatomic and physiologic laterality measures in anxious children may shed light on the treatment of comorbid mood and anxiety disorders in adults. In summary, pediatric GAD was associated with significantly larger STG total and gray and white matter volumes and a more pronounced right ⬎ left asymmetry in total and white matter STG volumes compared with sociodemographically similar control subjects. Prefrontal cortex and thalamus volumes did not differ between groups. Unlike findings in most neuropsychiatric disorders, the relevant brain region implicated in social processing and perhaps social anxiety and fear-related behaviors is larger in patients than in control subjects. These preliminary findings agree with previous work implicating posterior right-hemispheric regions in anxiety disorders and suggest that there are developmental alterations in the STG in pediatric GAD. Thus, dysmorphometry of the STG and the amygdala may represent a vulnerability to childhood GAD. Interpretation of our results are limited by the following. 1) We were not able to obtain detailed and extensive neuropsychologic testing in our subjects and thus do not know the functional significance of these findings. 2) We studied a small sample of GAD subjects. 3) The effect size of our STG findings is small but in the range of published significant findings for quantitative MRI studies (Lange et al 1997). Thus, the etiology, neuropsychologic function, specificity, and the persistence of these anatomic findings in pediatric GAD need to be examined. Future anatomic and functional MRI brain studies of childhood GAD and of children at risk for anxiety disorders targeting STG, amygdala, and other brain regions involved in processing social cues are warranted. This study was supported by NIMH Grant No. 5 K08 MHO1324-02 (principal investigator: MDDB), NIMH Grants Nos. MH01180 and MH43687 (principal investigator: MSK), NIMH Grant No. MH 41712 (“The Psychobiology of Anxiety and Depression in Children and Adolescents,” principal investigator: NDR), and a 1995 and 1998 NARSAD Young Investigators Award (principal investigator: MDDB).The authors thank the following individuals for their assistance in this work: Amy M. Boring, Karin Frustaci, and Laura Trubnick, RN. The authors thank Jay N. Giedd, M.D., and A. Catherine Vaituzis for their technical consultation concerning superior temporal gyrus measures and Andrew Gilbert, M.D., and Keith Harenski for their technical consultation concerning thalamus measures. We also thank Robert Sweet, M.D., and Daniel S. Pine, M.D., for their helpful comments.

References Baron-Cohen S, Ring HA, Wheelwright S, Bullmore ET, Brammer MJ, Simmons A, et al (1999): Social intelligence in the

Superior Temporal Gyrus Volumes in Pediatric GAD

normal and autistic brain; an fMRI study. Eur J Neurosci 11:1891–1898. Biederman J, Rosenbaum JF, Bolduc-Murphy EA, Faraone SV, Chaloff J, Hirshfeld DR, et al (1993): Behavioral inhibition as a temperamental risk factor for anxiety disorders. Child Adolesc Psychiatr Clin North Am 2:667– 683. Birmaher B, Brent DA, Chiappetta L, Bridge J, Monga S, Baugher M (1999): Psychometric Properties of the Screen for Child Anxiety Related Emotional Disorders Scale (SCARED): A Replication Study. J Am Acad Child Adolesc Psychiatry 38:1230 –1236. Birmaher B, Khetarpal S, Brent DA, Cully M, Balach L, Kaufman J, et al (1997): The Screen for Child Anxiety Related Emotional Disorders (SCARED): Scale Construction and Psychometric Characteristics. J Am Acad Child Adolesc Psychiatry 36:545–553. Brothers L (1990): The social brain: A project for integrating primate behavior and neurophysiology in a new domain. Concepts Neurosci 1:27–51. Bruder GE, Stewart JW, Tenke CE, McGrath PJ, Leite P, Bhattacharya N, et al (2001): Electroencephalographic and perceptual asymmetry differences between responders and nonresponders to an SSRI antidepressant. Biol Psychiatry 49:416 – 425. Caspi A, Moffitt TE, Newman DL, Silva PA (1996): Behavioral observations at age 3 years predict adult psychiatric disorders. Arch Gen Psychiatry 53:1033–1039. Chambers WJ, Puig-Antich J, Hirsch M, Paez P, Ambrosini PJ, Tabrizi MA, et al (1985): The Assessment of Affective Disorders in Children and Adolescents by Semi-Structured Interview: Test-retest reliability of the schedule for affective disorders and schizophrenia for school-age children, present episode version. Arch Gen Psychiatry 42:696 –702. Davidson RJ, Abercrombie H, Nitschke JB, Putnam K (1999): Regional brain function, emotion and disorders of emotion. Curr Opin Neurobiol 9:228 –234. De Bellis MD, Casey BJ, Dahl R, Birmaher B, Williamson D, Thomas KM, et al (2000): A pilot study of amygdala volumes in pediatric generalized anxiety disorder. Biol Psychiatry 48:51–57. De Bellis MD, Hall J, Boring AM, Frustaci K, Moritz G (2001a): A pilot longitudinal study of hippocampal volumes in pediatric maltreatment-related posttraumatic stress disorder. Biol Psychiatry 50:305–309. De Bellis MD, Keshavan M, Clark DB, Casey BJ, Giedd J, Boring AM, et al (1999): A.E. Bennett Research Award. Developmental Traumatology Part II: Brain Development. Biol Psychiatry 45:1271–1284. De Bellis MD, Keshavan M, Frustaci K, Shifflett H, Iyengar S, Beers SR, et al (2002): Superior temporal gyrus volumes in maltreated children and adolescents with PTSD. Biol Psychiatry 51:544 –552. De Bellis MD, Keshavan MS, Beers SR, Hall J, Frustaci K, Masalehdan A, et al (2001b): Sex differences in brain maturation during childhood and adolescence. Cereb Cortex 11:552–557. Denckla MB (1985): Revised physical and neurological examination for soft signs. Psychopharmacol Bull 21:773– 800. Desimone R (1991): Face-selective cells in the temporal cortex of monkeys. J Cogn Neurosci 3:1– 8.

BIOL PSYCHIATRY 2002;51:553–562

561

Etcoff NL (1986): The neuropsychology of emotional expression. In Goldstein G, Tarter RE, editors. Advances in Clinical Neuropsychology, Vol 3. New York: Plenum Press, 127–179. Giedd JN, Vaituzis AC, Hamburger SD, Lange N, Rajapakse JC, Kaysen D, et al (1996): Quantitative MRI of the temporal lobe, amygdala, and hippocampus in normal human development: Ages 4 –18. J Comp Neurol 366:223–230. Gilbert AR, Moore GJ, Keshavan MS, Paulson LA, Narula V, Mac Master FP, et al (2000): Decrease in thalamic volumes of pediatric patients with obsessive-compulsive disorder who are taking paroxetine. Arch Gen Psychiatry 57:449 – 456. Hasselmo ME, Roll s ET, Baylis GC (1989): The role of expression and identity in the face-selective responses of neurons in the temporal visual cortex of the monkey. Behav Brain Res 32:203–218. Heller W (1993): Neuropsychological mechanisms of individual differences in emotion, personality, and arousal. Neuropsychology 7:476 – 489. Heller W, Nitschke JB, Etienne MA, Miller GA (2000): Patterns of regional brain activity differentiate types of anxiety. J Abnorm Psychol 106:376 –385. Hollingshead AB (1975): Four Factor Index of Social Status. New Haven CT: Department of Sociology, Yale University. Jacobsen L, Giedd JN, Castellanos FX, Vaituzis CA, Hamburger SD, Kumra S, et al (1998): Progressive reduction of temporal lobe structures in childhood-onset schizophrenia. Am J Psychiatry 155:678 – 685. Kagan J, Snidman N (1999): Early childhood predictors of adult anxiety disorders. Biol Psychiatry 46:1536 –1541. Kalin NH, Larson C, Shelton SE, Davidson RJ (1998): Asymmetric frontal brain activity, cortisol, and behavior associated with fearful temperament in rhesus monkeys. Behav Neurosci 112:286 –292. Kaufman J, Birmaher B, Brent D, Rao U, Flynn C, Moreci P, et al (1997): Schedule for Affective Disorders and Schizophrenia for School-Age Children-Present and Lifetime Version (K-SADS-PL): Initial reliability and validity data. J Am Acad Child Adolesc Psychiatry 36:980 –988. Keller J, Nitschke JB, Bhargava T, Deldin PJ, Gergen JA, Miller GA, et al (2000): Neuropsychological differentiation of depression and anxiety. J Abnorm Psychol 109:3–10. Kentgen LM, Tenke CE, Pine DS, Fong R, Klein RG, Bruder GE (2000): Electroencephalographic asymmetries in adolescents with major depression: Influence of comorbidity with anxiety disorders. J Abnorm Psychol 109:797– 802. LaBar KS, Gatenby JC, Gore JC, LeDoux JE, Phelps EA (1998): Human amygdala activation during conditioned fear acquisition and extinction: A mixed-trial fMRI study. Neuron 20:937–945. Lange N, Giedd JN, Castellanos FX, Vaituzis AC, Rapoport JL (1997): Variability of human brain structure size: Ages 4 –20 years. Psychiatry Res 74:1–12. Lloyd RL, Kling AS (1991): Delta activity from amygdala in squirrel monkeys (saimiri sciureus): Influence of social and environmental contexts. Behav Neurosci 105:223–229. Orvaschel H, Puig-Antich (1987): Schedule for Affective Disorder and Schizophrenia for School-Age Children, Epidemiologic Version (K-SADS-E), 4th version. Pittsburgh, Pennsylvania: Western Psychiatric Institute and Clinic.

562

BIOL PSYCHIATRY 2002;51:553–562

Pine DS, Kentgen LM, Bruder GE, Leite P, Bearman K, Ma Y, et al (2000): Cerebral laterality in adolescent major depression. Psychiatry Res 93:135–144. Quirk GJ, Armony JL, LeDoux JE (1997): Fear conditioning enhances different temporal components of tone-evoked spike trains in auditory cortex and lateral amygdala. Neuron 19:613– 624. Rasband W (1996): NIH image manual. Bethesda, MD: National Institutes of Health. Roberts M, Hanaway J (1971): Atlas of the Human Brain in Section. Philadelphia, Pennsylvania: Lea & Febiger. Rosenberg DR, Averbach DH, O’Hearn KM, Seymour AB, Birmaher B, Sweeney JA (1997a): Oculomotor response inhibition abnormalities in pediatric obsessive-compulsive disorder. Arch Gen Psychiatry 54:831– 838. Rosenberg DR, Keshavan MS, O’Hearn KM, Dick EL, Bagwell WW, Seymour AB, et al (1997b): Frontostriatal measurement

M.D. De Bellis et al

in treatment-naive children with obsessive-compulsive disorder. Arch Gen Psychiatry 54:824 – 830. Shenton ME, Kikinis R, Jolesz FA, Pollak SD, LeMay M, Wible CG, et al (1992): Abnormalities of the left temporal lobe and thought disorder in schizophrenia. A quantitative magnetic resonance imaging study. New Engl J Med 327:604 – 612. Stapleton JM, Morgan MJ, Liu X, Yung BC, Phillips RL, Wong DF, et al (1997): Cerebral glucose utilization is reduced in second test session. J Cereb Blood Flow Metab 17:704 –712. Wechsler D (1974): Manual for the Wechsler Intelligence Scale for Children—Revised. San Antonio, TX: Psychological Corporation. Wu JC, Buchsbaum MS, Hershey TG, Hazlett E, Sicotte N, Johnson JC (1991): PET in generalized anxiety disorder. Biol Psychiatry 29:1181–1199. Yuh WTC, Tah ET, Afifi AK, Sahinoglu K, Gao F, Bergman RA (1994): MRI of Head and Neck Anatomy. New York: Churchill Livingstone.