Reduced size and abnormal asymmetry of parietal cortex in women with borderline personality disorder

Reduced Size and Abnormal Asymmetry of Parietal Cortex in Women with Borderline Personality Disorder Eva Irle, Claudia Lange, and Ulrich Sachsse Background: Evidence is accumulating that suggests borderline personality disorder (BPD) and posttraumatic stress disorder (PTSD) are related to small hippocampal size. Psychotic symptoms are frequent in both disorders. Psychotic spectrum disorders are known to be related to abnormalities of temporoparietal cortices. Methods: Using structural magnetic resonance imaging (3D-MRI), parietal cortex and hippocampal volumes were assessed in 30 young women with BPD who had been exposed to severe childhood sexual and physical abuse and in 25 healthy control subjects. Results: Compared with control subjects, BPD subjects had significantly smaller right parietal cortex (⫺11%) and hippocampal (⫺17%) volumes. The parietal cortex of borderline subjects showed a significantly stronger leftward asymmetry when compared with control subjects. Stronger psychotic symptoms and schizoid personality traits in borderline subjects were significantly related to reduced leftward asymmetry. Stronger trauma-related clinical symptoms and neuropsychologic deficits were significantly related to smaller hippocampal size. Conclusions: Our results are consistent with previous findings of small hippocampal size in BPD and PTSD. Reduced right parietal cortex size in individuals with BPD may reflect a neurodevelopmental deficit of the right hemisphere.

Key Words: Hippocampus, parietal asymmetry, posttraumatic stress disorder, psychotic symptoms, structural magnetic resonance imaging (3D-MRI), trauma

B

orderline personality disorder (BPD) is defined as an intermediate level of personality organization that is considered to occupy a borderline area between neurosis and psychosis (Kernberg 1967; Kernberg et al 2000). Stressrelated dissociative and psychotic symptoms occur in about 75% of individuals with BPD (Skodol et al 2002) and in about 40% of individuals with posttraumatic stress disorder (PTSD; David et al 1999; Hamner et al 1999). Psychotic symptoms in BPD and PTSD usually involve nonbizarre hallucinations or delusions and are not associated with formal thought disorder or flat or inappropriate affect. They tend to be transient (lasting minutes or hours), and reality testing is usually impaired for only a short time. Psychotic symptoms in PTSD may cause substantial psychologic distress (Hamner 1997) and are associated with higher levels of corticotropin-releasing factor (Sautter et al 2003). They seem not to be due to reexperience of traumatic events (Butler et al 1996; Hamner et al 1999). Individuals with BPD or PTSD do not have an increased prevalence of psychotic-spectrum disorders (Breslau et al 1991; Zanarini et al 1998), as is true for their families (Sautter et al 2002; White et al 2003). Research thus far has focused on size reductions of the amygdala and hippocampus in individuals with BPD who were exposed to childhood physical or sexual abuse (Driessen et al 2000; Rüsch et al 2003; Schmahl et al 2003a; Tebartz van Elst et al 2003). Functional imaging studies show that persons with BPD display pronounced prefrontal dysfunction (de la Fuente et al 1997; Juengling et al 2003; Soloff et al 2000, 2003) and enhanced prefrontal and amygdala activation in response to emotional stimuli (Donegan et al 2003; Herpetz et al 2001; From the Department of Psychiatry and Psychotherapy (EI, CL), University of Göttingen, and Psychiatric State Hospital of Lower Saxony (US), Göttingen, Germany. Address reprint requests to Dr. Eva Irle, Department of Psychiatry and Psychotherapy, University of Göttingen, Von-Siebold-Str. 5, D-37075 Göttingen, Germany; E-mail: [email protected]. Received March 5, 2004; revised July 14, 2004; accepted October 2, 2004.

0006-3223/05/$30.00 doi:10.1016/j.biopsych.2004.10.004

Schmahl et al 2003b). Dysfunction in prefrontal regions (Brodmann areas 9 –11) is suggested to be implicated in the deficits of persons with BPD to regulate emotional behavior (Soloff et al 2003). Studies investigating the structural and functional neural correlates of psychotic features in BPD and PTSD are lacking. However, research on individuals with schizophrenia has repeatedly pointed to abnormalities of temporoparietal cortices. Individuals with schizophrenia were shown to have a reduced leftward asymmetry, size reduction, or both refers to all anatomical structures (Shapleske et al 1999); inferior parietal cortex (Frederikse et al 2000; Niznikiewicz et al 2000); and perisylvian cortex (Sommer et al 2001). Reduced leftward asymmetry of the Heschl gyrus was also demonstrated for persons with schizotypal personality disorder (Dickey et al 2002). Size reductions of left-sided temporoparietal cortices were shown to be associated with psychotic symptoms of individuals with schizophrenia, indicating more severe symptoms in individuals with more pronounced size reductions (Barta et al 1990; Gaser et al 2004; Kwon et al 1999; Shenton et al 1992). A recent study demonstrated that transcranial magnetic stimulation of the left temporoparietal cortex improves auditory hallucinations in schizophrenia (Hoffman et al 2003). Increasing evidence suggests that glutamatergic dysfunction represents an important part of the pathophysiology of psychotic states. The subanesthetic application of the N-methyl-D-aspartate (NMDA) antagonist ketamine in humans is known to produce psychotic and dissociative symptoms (Krystal et al 1994). It has been proposed that NMDA receptor hypofunction might cause excitotoxic limbic (i.e., hippocampal) and temporoparietal cortical neurodegeneration in schizophrenia (Olney and Farber 1995). Evidence is increasing that stress-related hyperglutamatergic states may also contribute to dissociative symptoms and neural toxicity (i.e., hippocampal degeneration) in individuals who have been exposed to traumatic stress (Chambers et al 1999). A recent study found that burn victims with enduring ketamine application in the posttraumatic state showed significantly stronger PTSD symptoms than burn victims without such treatment (Winter and Irle 2004). More severe PTSD symptoms were related to smaller hippocampal size, as previous investigations have shown (Bremner et al 1997, 2003; Gilbertson et al 2002; Gurvits et al 1996; Villarreal et al 2002). BIOL PSYCHIATRY 2005;57:173–182 © 2005 Society of Biological Psychiatry

174 BIOL PSYCHIATRY 2005;57:173–182

E. Irle et al

Table 1. Demographic and Clinical Characteristics of Borderline and Control Subjects Variablea Age (years) Education (years) Height (cm) Handedness (left:right)b Duration of Trauma-Related Clinical Symptoms (years) Traumatic Antecedent Questionnairec Dissociative Experiences Scale (total score) Diagnostic Interview for Borderlines, Psychotic Symptoms SCID-I, Psychotic Symptoms SCID-II, Schizoid Traits SCID-II, Schizotypal Traits Impact of Events Scale-Revised Intrusions Avoidance Hyperarousal Beck Depression Inventory Münchner Alkoholismustestd

Borderline Subjects (n ⫽ 30)

Control Subjects (n ⫽ 25)

31 ⫾ 6 11 ⫾ 2 168 ⫾ 6 2:28 14 ⫾ 7 4.8 ⫾ 1.2 24.7 ⫾ 12.3 5.5 ⫾ 3.0 1.7 ⫾ 2.3 1.5 ⫾ 2.1 1.7 ⫾ 1.6

33 ⫾ 7 11 ⫾ 2 169 ⫾ 6 0:25 2.4 ⫾ .5 2.1 ⫾ .4

Statistic t(53) ⫽ ⫺.89 t(53) ⫽ .23 t(53) ⫽ ⫺.21 t(29.1) ⫽ 8.90 t(50) ⫽ 8.17

p .373 .822 .836 .495 .000 .000

3.0 ⫾ 1.2 3.7 ⫾ 1.0 3.1 ⫾ 1.3 27 ⫾ 9 6⫾7

SCID, Structured Clinical Interview for DSM-IV. Table values are given as mean ⫾ SD unless otherwise indicated. b Fisher’s Exact Test. c Neglect and physical and sexual abuse experiences during childhood and adolescence (7–18 years). Six borderline subjects were unable to complete parts or all of the questionnaire. d Clinician-administered rating scale; 0 –5 ⫽ normal alcohol consumption, 6 –10 ⫽ likely alcohol abuse, 11–52 ⫽ alcohol abuse or dependence. a

In this investigation, the volumes of the parietal cortex and the hippocampus of 30 women with borderline personality disorder (BPD) who had been exposed to severe childhood sexual and physical abuse were compared with those of a healthy matched control group (n ⫽ 25). The goals of our study were 1) to investigate whether parietal cortex size is reduced in a sample of BPD subjects, 2) to investigate whether the parietal cortex of BPD subjects has an abnormal asymmetry, 3) to analyze how psychotic symptoms and psychosis-related personality traits are associated with parietal cortex size and asymmetry, and 4) to verify whether hippocampal size is reduced in BPD subjects and is associated with trauma-related symptoms.

Methods and Materials Subjects The sample comprised 30 young female inpatients with the diagnosis of borderline personality disorder (BPD) according to the Diagnostic and Statistical Manual of Mental Disorders (DSM-IV; American Psychiatric Association 1994) admitted to the Psychiatric State Hospital of Lower Saxony, Göttingen, Germany (Table 1). The hospital has a specialized therapeutic unit for women who have experienced severe childhood sexual and physical abuse. All subjects were assessed with a routine physical and neurologic examination, magnetic resonance imaging, and laboratory testing. Subjects with a history of neurologic disease (multiple sclerosis; n ⫽ 1) or electroencephalograph abnormalities being indicative for temporal lobe epilepsy (n ⫽ 2) or hyperintense magnetic resonance imaging (MRI) signals (n ⫽ 2) were excluded beforehand. Subjects with psychotic disorders (DSM-IV Axis I) were also excluded. Eight subjects were on antidepressant medication. Some subjects were occasionally treated with benzodiazepines (n ⫽ 5) or mild neuroleptics (n ⫽ 6). All borderline subjects were investigated with the Structured Clinical Interview for DSM-IV (SCID-I and SCID-II; First et al www.elsevier.com/locate/biopsych

1995, 1997; Wittchen et al 1997) and the Structured Clinical Interview for DSM-IV Dissociative Disorders (SCID-D; Gast et al 2000; Steinberg 1994) conducted by a master’s-level psychologist. Seven subjects received a second SCID-I interview conducted by another clinician. Both interviewers were blind to each other but had access to clinical information. Agreement between interviewers (expressed as kappa, a statistic that corrects for chance agreement) was 1.0 (p ⫽ .008) for the current or lifetime diagnosis of major depression and PTSD, respectively. All borderline subjects met DSM-IV criteria for BPD. All subjects but one had received the diagnosis of BPD by clinical judgment at an earlier time. Four subjects met DSM-IV criteria for schizoid personality disorder, five for paranoid personality disorder and two for schizotypal personality disorder. Eleven borderline subjects (37%) met criteria for lifetime or current PTSD. Three (10%) subjects met criteria for lifetime panic disorder with agoraphobia, eight (27%) for current panic disorder with agoraphobia, one subject (3%) for lifetime agoraphobia without panic disorder, and three (10%) for current agoraphobia without panic disorder. Three (10%) subjects met criteria for current generalized anxiety disorder, two (7%) for lifetime social phobia, six (20%) for current social phobia, and six (20%) for current obsessive– compulsive disorder. Thirteen (45%) subjects met criteria for lifetime major depression and 18 (60%) for current major depression. Three (10%) subjects met criteria for current somatization disorder and three (10%) for current pain disorder. Six (20%) subjects met criteria for lifetime anorexia, three (10%) for current anorexia, three (10%) for lifetime bulimia, and seven (23%) for current bulimia. Six (20%) subjects met criteria for lifetime alcohol abuse, 2 (7%) for current alcohol abuse, three (10%) for lifetime alcohol dependence, and two (7%) for lifetime sedative abuse. Four (14%) subjects met criteria for lifetime depersonalization disorder and 23 (77%) for current depersonalization disorder. One (3%) subject met criteria for lifetime dissociative amnesia, six (20%) for current dissociative amnesia, and four (13%) for current dissociative identity disorder.

E. Irle et al

BIOL PSYCHIATRY 2005;57:173–182 175

Figure 1. Boundaries defining the parietal cortex on the lateral (left) and medial (right) brain surface. Sulcal boundaries are labeled with black lines. (A) Posterior extension of the line passing through the sylvian fissure serving as postsylvian ventrolateral boundary. (B) Vertical line connecting the central and cingulate sulcus serving as anteriomedial boundary.

Borderline subjects were compared with 25 healthy female control subjects comparable with regard to age, height, and years of education. Control subjects were recruited for the study by an advertisement in a local newspaper and leaflets distributed in the Hospital of the University of Göttingen and in town. Only subjects without a history of neurologic or psychiatric (as assessed by the SCID-I) disorder were studied. After a complete description of the study was given to the subjects, written informed consent was obtained. The study design was approved by the Ethical Committee of the Medical Faculty of the University of Göttingen. Clinical and Neuropsychologic Assessment The Traumatic Antecedent Questionnaire (TAQ; Herman et al 1989) was used to assess neglect as well as physical and sexual abuse experiences during childhood and adolescence (7–18 years). The Impact of Events Scale-Revised (IES-R; Maercker and Schützwohl 1998; Weiss and Marmar 1996) was used to assess PTSD symptoms. Duration of trauma-related clinical symptoms was defined as the number of years since PTSD symptoms or other trauma-related clinical symptoms appeared. Delusions and hallucinations were assessed by aid of the SCID-I (screening module for psychotic symptoms; score: 0 –15). The Diagnostic Interview for Borderline Patients (DIB; Gunderson and Kolb 1981; Pütterich 1990) was used to assess a broader spectrum of psychotic symptoms (i.e., paranoid ideation, odd beliefs, unusual perceptual experiences, depersonalization and derealization, delusions, hallucinations, and psychotic symptoms in association with substance use; score: 0 –16). DSM-IV cluster A personality traits (i.e., fulfilled criteria of paranoid, schizoid and schizotypal personality disorder, respectively) were assessed by the SCID-II. The Dissociative Experiences Scale (DES; Bernstein and Putnam 1986; Freyberger et al 1999) was applied as a measure of dissociative symptom severity. Depressive symptoms were assessed by the Beck Depression Inventory (BDI; Beck et al 1960; Hautzinger et al 1995). The amount of alcohol consumption and alcohol-related clinical symptoms were assessed by the Münchner Alkoholismus-Test (MALT) (Feuerlein et al 1979). Intellectual, mnemonic, and attentional functions were assessed by the Wechsler Adult Intelligence Scale—Revised (WAIS-R; Tewes 1991; Wechsler 1981), the Wechsler Memory Scale—Revised (WMS-R; Wechsler 1987), and a computer-driven working memory test (Testbatterie zur Aufmerksamkeitsprüfung [TAP]; Zimmermann and Fimm 1993).

MRI Acquisition and Analysis All subjects received an MRI scan using a 1.5-T Philips Gyroscan machine. Scanning parameters of the T1-weighted three-dimensional sequence were as follows: echo time ⫽ 6.0 msec; repetition time ⫽ 24.0 ms; flip angle ⫽ 30°; field of view ⫽ 256; slice plane ⫽ sagittal; matrix ⫽ 256 ⫻ 256; slice thickness ⫽ 1.3 mm; slice number ⫽ 130; acquisition mode ⫽ three-dimensional (3D). Volumetric analysis was done on the basis of 3D MRIs. The images were transferred to a computer workstation and processed using the CURRY software (version 4.5; Compumedics, Melbourne, Australia). Images were reformatted into continuous 1-mm-thick slices. Intracranial volume and total brain volume was calculated with automated multistep algorithms and 3D region growing methods that are limited by gray-value thresholds. Simultaneous 3D visualization of brain structures and manual tracing allowed a precise identification and delineation of regions of interest. We used sulcal landmarks for the delineation of the parietal cortex because they have the advantage of directly modeling the substantial interindividual morphologic variability of the human cerebral cortex (Rademacher et al 1992). Parietal Cortex. The anterior border of the parietal cortex was defined by the central sulcus. On the medial surface, a vertical line was drawn between central and cingulate sulcus. The posterior border was defined by the parietooccipital sulcus. The ventrolateral border was defined by the sylvian fissure. A line passing through the sylvian fissure was extended posteriorly and served as postsylvian ventrolateral boundary. The inferior medial boundary was defined by a line passing through the cingulate sulcus, which was extended into the parietooccipital sulcus. Manual markings were made on coronal slices by drawing a straight line between sylvian fissure and central sulcus and between the posterior extension of the sylvian fissure and the cingulate sulcus. The resulting volume (Figure 1) includes gray and white matter of the postcentral, supramarginal, and angular gyri; the superior parietal lobe; and the precuneus (Brodmann areas 1–3, 5, 7, 39, and 40). Parietal cortex asymmetry was calculated as follows: (left volume – right volume) / (.5 * [left volume ⫹ right volume]). Positive values indicate left-greater-than-right asymmetry, and negative values indicate right-greater-than-left asymmetry. Hippocampus. The hippocampus was disarticulated from surrounding tissue on coronal slices by means of manual tracing according to a standardized protocol (Pruessner et al 2000) and www.elsevier.com/locate/biopsych

176 BIOL PSYCHIATRY 2005;57:173–182

E. Irle et al

Table 2. Brain Measures of Borderline and Control Subjects Volume (mL)a Intracranial Volume Total Brain Parietal Cortexc Total Lefte Righte Asymmetry Indexf Hippocampusd Total Lefte Righte

Borderline Subjects (n ⫽ 30) 1363 ⫾ 97 1074 ⫾ 99

Control Subjects (n ⫽ 25) 1384 ⫾ 88 1106 ⫾ 81

Difference (%) and Effect Sizeb

Statistic

p

⫺1.5 (⫺.23) ⫺2.9 (⫺.36)

t(53) ⫽ ⫺0.79 t(53) ⫽ ⫺1.27

.432 .209

T ⫽ 4.43 T ⫽ 1.09 T ⫽ 7.92 t(53) ⫽ 2.07

.035 .296 .005 .043

130 ⫾ 20 68 ⫾ 10 63 ⫾ 10 .069 ⫾ .081

142 ⫾ 19 72 ⫾ 10 70 ⫾ 9 .024 ⫾ .082

⫺9.2 (⫺.62) ⫺5.9 (⫺.40) ⫺11.1 (⫺.74)

5.3 ⫾ 0.8 2.6 ⫾ 0.5 2.7 ⫾ 0.4

6.2 ⫾ 0.7 3.0 ⫾ 0.4 3.2 ⫾ 0.3

⫺17.0 (⫺1.20) ⫺15.4 (⫺.89) ⫺18.5 (⫺1.43)

T ⫽ 18.92 T ⫽ 11.07 T ⫽ 19.67

.000 .008 .000

Table values are given as mean ⫾ SD. Percentage of difference in regional volumes is relative to control subjects. Effect sizes (in parentheses) are group mean differences divided by pooled SD. c The hypothesis-driven group ⫻ hemisphere (2 ⫻ 2) analysis of covariance (ANCOVA)-type statistic using total brain volume as covariate had the following T values group, T(1) ⫽ 4.43, p ⫽ .035; hemisphere, T(1) ⫽ 14.27, p ⫽ .000; group ⫻ hemisphere, T(1) ⫽ 4.93, p ⫽ .026. d The hypothesis-driven group ⫻ hemisphere (2 ⫻ 2) ANCOVA-type statistic using total brain volume as covariate had the following. T values: group, T(1) ⫽ 18.92, p ⫽ .000; hemisphere, T(1) ⫽ 15.16, p ⫽ .000; group ⫻ hemisphere, T(1) ⫽ 1.03, p ⫽ .310. e One-way ANCOVA-type statistic controlling for total brain volume. Box-approximation df ⫽ 1. f (Left volume - right volume)/(.5 * (left volume ⫹ right volume)). Positive values indicate left-greater-than-right asymmetry. a

b

with the aid of the serial sections provided by Duvernoy (1998). The anterior border of the hippocampus was found on the coronal slice showing the alveus, the uncal recess of the inferior horn of the lateral ventricle, or both, and the posterior border was found on the slice where an ovoid mass of gray matter appeared inferiomedially to the trigone of the lateral ventricle. All analyses were done blind to subjects’ test performance. For defining the intrarater and interrater reliability (the raters were blind to the diagnosis), each hemisphere of 10 randomly chosen cases were reassessed, respectively. The intraclass correlation coefficients for this procedure were r ⫽ .98 (intrarater) and r ⫽ .98 (interrater) for the parietal cortex and r ⫽ .95 (intrarater) and r ⫽ .96 (interrater) for the hippocampus. Statistical Analysis Student t tests (or Mann–Whitney U tests in case of small subject numbers) were applied to compare differences between groups on demographic, clinical, and neuropsychologic variables and on intracranial and total brain volume. Frequencies were compared using Fisher’s Exact Test. Partial correlation coefficients controlling for total brain volume and multiple regression analyses were used to examine the relationship between regional brain volumes and clinical and neuropsychologic variables. The alpha of correlation and regression analyses was not adjusted because the data are considered exploratory. We used nonparametric tests to analyze regional brain volumes because of deviations of left and right parietal cortex volumes from normal distribution in both groups of subjects (Kolmogorov–Smirnov test; D values between .155 and .186, p values between .01 and .09). The parietal cortex and the hippocampus of borderline and control subjects was compared by nonparametric 2 ⫻ 2 analyses of covariance (ANCOVA; Brunner et al 2002) controlling for total brain volume with the betweensubjects factor of Group (borderline, control subjects) and the within-subjects factors of Hemisphere (left, right). All analyses were two-tailed, and alpha was defined as p ⬍ .05. Statistical computations were performed using the Statistical Package for the Social Sciences (SPSS for Windows, version 10.0) and the Statistical Analysis System (SAS for Windows, verwww.elsevier.com/locate/biopsych

sion 8.01; nonparametric ANCOVAs: http//www.ams.med. uni-goettingen.de/Projekte/LD/Makros_LD.html).

Results Brain Measures: Group Comparisons The groups did not differ with respect to intracranial volume and total brain volume (Table 2). The 2 (Group) ⫻ 2 (Hemisphere) ANCOVA comparing the parietal cortex of borderline and control subjects yielded a significant effect of Group (p ⫽ .035) and Hemisphere (p ⫽ .000), indicating larger parietal cortex volumes of control subjects and larger left compared with right parietal cortex volumes (Figure 2). The interaction of Group and Hemisphere was also significant (p ⫽ .026). The follow-up regional ANCOVA models revealed that borderline subjects had significantly smaller right parietal cortices (p ⫽ .005) compared with control subjects.

Figure 2. Scatter plot of left and right parietal cortex volumes (mL) for subjects with borderline personality disorder and healthy controls. Group means are indicated by horizontal lines.

E. Irle et al

Figure 3. Scatter plot of parietal cortex asymmetry values of subjects with borderline personality disorder and healthy control subjects. Group means are indicated by horizontal lines. Positive values indicate left-greater-thanright asymmetry; and negative values indicate right-greater-than-left asymmetry.

Parietal cortex asymmetry was significantly different (p ⫽ .043) between borderline and control subjects, indicating a more pronounced left-greater-than-right asymmetry in borderline compared with control subjects (Figure 3). Parietal cortex asymmetry values of two left-handed borderline subjects (see Table 1) were within .5 SD of those of the whole BPD sample. The 2 (Group) ⫻ 2 (Hemisphere) ANCOVA comparing the hippocampus of borderline and control subjects yielded a significant effect of Group (p ⫽ .000) and Hemisphere (p ⫽ .000), indicating smaller hippocampal volumes of borderline compared with control subjects and larger right compared with left hippocampal volumes (Figure 4). For detailed results of all statistical comparisons, see Table 2.

Figure 4. Scatter plot of left and right hippocampal volumes (mL) for subjects with borderline personality disorder and healthy control subjects. Group means are indicated by horizontal lines.

BIOL PSYCHIATRY 2005;57:173–182 177

Figure 5. Relationship between parietal cortex asymmetry and number of psychotic symptoms as assessed by the screening module for psychotic symptoms (Structured Clinical Interview for DSM-IV—I; see Clinical and Neuropsychological Assessment in Methods and Materials; rs ⫽ –.41; p ⫽ .026). The best-fitting line has an exponential function (R2 ⫽ .15; p ⫽ .036). Positive values on the x axis indicate left-greater-than-right asymmetry; negative values indicate right-greater-than-left asymmetry. Asymmetry values of two subjects coincided exactly in two cases. For comparison, 15 female subjects with schizophrenia from the study of Frederikse et al (2000) presented with an inferior parietal cortex asymmetry of ⫺.11 ⫾ .28.

Relationship Between Psychotic Symptoms and Parietal Cortex Measures All borderline subjects reported frequent episodes of depersonalization and derealization (DIB). Nineteen borderline subjects experienced nonbizarre hallucinations or delusions as assessed by the screening module for psychotic symptoms (SCID-I). The content of delusions included referential, persecutory, somatic, or grandiose themes. Hallucinations occurred in all sensory modalities (i.e., auditory, visual, tactile, olfactory, or gustatory). In three subjects, auditory hallucinations included hearing voices; two of these subjects met criteria for current dissociative identity disorder. None of the subjects experienced disorganized or catatonic behavior. Subjects with psychotic symptoms according to the SCID-I (n ⫽ 19) did not differ significantly from subjects without psychotic symptoms (n ⫽ 11) on any clinical or neuropsychologic variable or on brain measures except for parietal cortex asymmetry (U test; p ⫽ .023), for which subjects with psychotic symptoms showed a weaker leftward asymmetry (.046 ⫾ .077) than subjects without psychotic symptoms (.109 ⫾ .076). Accordingly, the amount of psychotic symptoms (screening module for psychotic symptoms, SCID-I) was significantly related to parietal cortex asymmetry of borderline subjects (n ⫽ 30; rs ⫽ ⫺.41, p ⫽ .026), indicating a reduced leftward asymmetry in subjects with more psychotic symptoms (Figure 5). A very similar relationship was obtained for the amount of psychotic symptoms as assessed by the DIB (rs ⫽ ⫺.37; p ⫽ .050; Figure 6). Borderline subjects with a reduced leftward asymmetry displayed also more schizoid personality traits (SCID-II; rs ⫽ ⫺.59; p ⫽ .001) (Figure 7). Nonparametric partial correlations controlling for total brain volume revealed a positive relationship between right parietal cortex volume and psychotic symptoms (DIB; r ⫽ .43; p ⫽ .024), indicating larger right parietal cortex volumes in subjects with stronger psychotic symptoms. The relationship between right parietal cortex volume and schizotypal personality traits (SCIDII) closely approached conventional levels of significance (r ⫽ .36; p ⫽ .053). www.elsevier.com/locate/biopsych

178 BIOL PSYCHIATRY 2005;57:173–182

E. Irle et al values ⬍ .05), indicating better performance of subjects with larger parietal cortex volumes. For detailed results, see Table 3.

Figure 6. Relationship between parietal cortex asymmetry and number of psychotic symptoms as assessed by the Diagnostic Interview for Borderline Patients (see Clinical and Neuropsychologic Assessment in Methods and Materials; rs ⫽ ⫺.37; p ⫽ .050). Positive values on the x axis indicate leftgreater-than-right asymmetry; negative values indicate right-greater-thanleft asymmetry.

Relationship Between Trauma-Related Symptoms and Hippocampal Volume Partial correlations controlling for total brain volume yielded a significant relationship between intrusions (IES-R) and right hippocampal volume (r ⫽ ⫺.56; p ⫽ .003), and between hyperarousal (IES-R) and left (r ⫽ ⫺.39; p ⫽ .047) and right (r ⫽ ⫺.44; p ⫽ .026) hippocampal volume, indicating more intrusions and stronger hyperarousal in subjects with smaller hippocampal volumes. The duration of trauma-related clinical symptoms was significantly related to left hippocampal volume (r ⫽ ⫺.54; p ⫽ .002), indicating smaller left hippocampal volumes in subjects with longer duration of symptoms. Stronger neglect (TAQ) during childhood and adolescence was significantly related to smaller left (rs ⫽ ⫺.57; p ⫽ .002) and right (rs ⫽ ⫺.43; p ⫽ .024) hippocampal volumes. There was also a significant negative relationship between the amount of neglect and total brain volume (rs ⫽ ⫺.46; p ⫽ .015). Partial correlations controlling for total brain volume between neglect and hippocampal volume were marginally significant for the left hippocampus (r ⫽ ⫺.38; p ⫽ .052) and insignificant for the right hippocampus. Neuropsychological Performance and Brain Measures Compared with control subjects, borderline subjects showed significantly reduced intellectual functions (WAIS-R). Borderline subjects were also impaired on tests of memory and attention (WMS-R; Table 3). The volumes of left and right hippocampus and left and right parietal cortex were entered into multiple regression analyses (method: stepwise; significance level for selecting variables: ␣ ⫽ .05). Total brain volume was entered as a further predictor to control for the influence of brain size. Considering borderline subjects, only right hippocampal volume significantly predicted intellectual, mnemonic, and attentional performance (proportion of explained variance (R2) between .15 and .30; p values ⬍ .05), and the addition of the other variables did not significantly improve the prediction. Larger right hippocampal volumes were related to better test performance. Considering control subjects, parietal cortex volume significantly predicted attentional test performance (R2 between .19 and .28; p www.elsevier.com/locate/biopsych

Influence of Comorbid Disorders and Medication Comparing borderline subjects fulfilling DSM-IV criteria for current or lifetime PTSD (n ⫽ 11) with borderline subjects not fulfilling PTSD criteria (n ⫽ 19) revealed smaller intracranial (1314 ⫾ 74 vs. 1389 ⫾ 99; U test, p ⫽ .027) and total brain (1021 ⫾ 68 vs. 1105 ⫾ 103; p ⫽ .011) volumes of subjects with PTSD. One-way ANCOVAs controlling for total brain volume revealed that subjects with and without PTSD did not differ with regard to left (2.4 ⫾ .4 vs. 2.7 ⫾ .5) and right (2.5 ⫾ .4 vs. 2.8 ⫾ .5) hippocampal volume and left (67 ⫾ 11 vs. 69 ⫾ 10) and right (62 ⫾ 11 vs. 64 ⫾ 10) parietal cortex volume (p values ⬎ .60). Parietal cortex asymmetry also did not differ significantly (.064 ⫾ .077 vs. .072 ⫾ .086; p ⫽ .767). Subjects with PTSD performed worse than subjects without PTSD on neuropsychologic measures (Verbal IQ, Verbal Memory, Visual Memory, Delayed Recall, Attention/Concentration; p values ⬍ .05) and reported more intrusions (IES-R; p ⫽ .001), more hyperarousal (IES-R; p ⫽ .003), and more severe sexual abuse (TAQ; p ⫽ .023) than subjects without PTSD. Borderline subjects with current or lifetime anorexia (n ⫽ 8) did not differ from subjects without anorexia (n ⫽ 22) with regard to intracranial volume (1329 ⫾ 95 vs. 1376 ⫾ 97), total brain volume (1024 ⫾ 112 vs. 1092 ⫾ 90), or parietal cortex asymmetry (.1024 ⫾ .084 vs. .057 ⫾ .079; U tests; p ⬎ .15), or with regard to left (2.5 ⫾ .6 vs. 2.6 ⫾ .4) and right (2.7 ⫾ .4 vs. 2.7 ⫾ .5) hippocampal volume or left (68 ⫾ 13 vs. 68 ⫾ 10) and right (61 ⫾ 11 vs. 64 ⫾ 10) parietal cortex volume (one-way ANCOVAs controlling for total brain volume; p ⬎ .50). Subjects with anorexia did not differ from subjects without anorexia on neuropsychologic measures (p ⬎ .20) except Working Memory (TAP), for which subjects with anorexia reacted slower than subjects without anorexia (795 ⫾ 144 vs. 623 ⫾ 157; p ⫽ .017). Borderline subjects with current or lifetime alcohol abuse or dependence (n ⫽ 10) did not differ from subjects without these disorders (n ⫽ 20) on brain volume measures or on parietal

Figure 7. Relationship between parietal cortex asymmetry and number of schizoid personality traits (criteria for schizoid personality disorder as assessed by the Structured Clinical Interview for DSM-IV—II; rs ⫽ ⫺.59; p ⫽ .001). The best-fitting line has an exponential function (R2 ⫽ .30; p ⫽ .002). Positive values on the x axis indicate left-greater-than-right asymmetry; and negative values indicate right-greater-than-left asymmetry. Asymmetry values of two subjects coincided exactly in one case.

BIOL PSYCHIATRY 2005;57:173–182 179

E. Irle et al Table 3. Neuropsychological Results of Borderline and Control Subjects Variablea Intelligence WAIS-R: Verbal IQ WAIS-R: Performance IQ WAIS-R: Block Designb Memory WMS-R: Verbal Memory WMS-R: Visual Memory WMS-R: Delayed Recall Attention/Working Memory WMS-R: Attention/Concentration WMS-R: Digit Span, fwdb WMS-R: Digit Span, bwdb WMS-R: Visual Span, fwdb WMS-R: Visual Span, bwdb TAP: Working Memory, msecc

Borderline Subjects (n ⫽ 30)

Control Subjects (n ⫽ 25)

Statistic

p

102 ⫾ 13 100 ⫾ 15e 30 ⫾ 9d

111 ⫾ 13 116 ⫾ 16 36 ⫾ 7

t(53) ⫽ ⫺2.53 t(53) ⫽ ⫺3.72 t(53) ⫽ ⫺2.49

.015 .000 .016

108 ⫾ 12d 101 ⫾ 18d 111 ⫾ 16d

112 ⫾ 14 118 ⫾ 17 119 ⫾ 14

t(53) ⫽ ⫺1.04 t(53) ⫽ ⫺3.53 t(53) ⫽ ⫺1.92

.302 .001 .061

91 ⫾ 15e 7.2 ⫾ 1.8d 6.4 ⫾ 2.1 8.2 ⫾ 1.4 7.7 ⫾ 1.8d 666 ⫾ 170

102 ⫾ 12f 8.0 ⫾ 1.7 7.0 ⫾ 2.0f 9.6 ⫾ 1.5 8.4 ⫾ 1.8 654 ⫾ 203g

t(53) ⫽ ⫺2.88 t(53) ⫽ ⫺1.52 t(53) ⫽ ⫺1.14 t(53) ⫽ ⫺3.37 t(53) ⫽ ⫺1.61 t(53) ⫽ .22

.006 .133 .259 .001 .113 .826

WAIS-R, Wechsler Adult Intelligence Scale-Revised (IQ estimates were derived from Information, Similarities, Picture Completion, and Block Design score); WMS-R, Wechsler Memory Scale-Revised; TAP, Testbatterie zur-Aufmerksamkeitsprufung. a Table values are given as mean ⫾ SD. b Raw scores. c Identical (one-figure) numbers appearing next to the last on a computer screen have to be identified by pressing a button. d Variable is significantly predicted by right hippocampal volume (p ⬍ .05). e Variable is significantly predicted by right hippocampal volume (p ⬍ .01). f Variable is significantly predicted by right parietal cortex volume (p ⬍ .05). g Variable is significantly predicted by left parietal cortex volume (p ⬍ .01). For further explanation, see Results, Neuropsychological Performance and Brain Measures.

cortex asymmetry (U tests or ANCOVAs controlling for total brain volume; p ⬎ .30). Relationships between alcohol consumption and alcohol-related clinical symptoms (MALT) and any brain measure were not significant for the whole BPD sample or BPD subjects with alcohol disorders, respectively (Spearman correlations; p ⬎ .10). Depressive symptom severity (BDI) was not related to intracranial or total brain volume, left or right hippocampal volume, left or right parietal cortex volume, or parietal cortex asymmetry (Spearman correlations; p ⬎ .40). Depressive symptom severity was also not related to any neuropsychologic measure (p ⬎ .10). Brain measures did not differ between borderline subjects taking antidepressant drugs (n ⫽ 8) and those who did not (n ⫽ 22; p ⬎ .30). Subjects taking benzodiazepines (n ⫽ 5) did not differ from those who did not (n ⫽ 25; p ⬎ .30), and subjects taking mild neuroleptics (n ⫽ 6) did not differ from those who did not (n ⫽ 24) (p-values ⬎ .40). Brain measures of subjects taking antidepressants, benzodiazepines or neuroleptics were within .5 SD of the brain measures of unmedicated subjects, respectively.

Discussion Reduced Size and Abnormal Asymmetry of the Parietal Cortex in BPD Our results demonstrate reduced size of the right parietal cortex in borderline subjects (Table 2, Figure 2). Accordingly, our data revealed a stronger parietal leftward asymmetry of borderline compared with control subjects (Figure 3). Psychotic symptoms and schizoid personality traits of borderline subjects were significantly related to parietal cortex asymmetry, indicating a reduced leftward asymmetry in subjects with stronger psychotic symptoms and more schizoid personality traits (Figures 5–7). Parietal cortex asymmetry of borderline subjects with pro-

nounced psychotic symptoms was similar to that reported for women with schizophrenia (Frederikse et al 2000; cf. Figure 5). Research on individuals with schizophrenia has demonstrated size reduction of the left temporoparietal cortices, reduced leftward asymmetry of these cortices when compared with control subjects, or both (Frederikse et al 2000; Niznikiewicz et al 2000; Shapleske et al 1999; Sommer et al 2001). The subgroup of our borderline subjects with pronounced psychotic and schizoid symptoms did also present with a reduced leftward asymmetry (see Figures 5–7). It might be speculated that a reduction of the normal left-greater-than-right asymmetry of temporoparietal cortices increases the likeliness for developing psychotic symptoms. Nevertheless, the majority of our borderline subjects had an increased leftward asymmetry due to reduced right parietal cortex size. Possibly, increased leftward parietal cortex asymmetry protects the majority of individuals with BPD from developing more pronounced and disabling psychotic symptoms. Currently, reduced leftward or even reversed temporoparietal asymmetry in schizophrenia is considered a neurodevelopmental deficit of the left hemisphere (Pearlson et al 1996; Shenton et al 2001). The pattern of parietal cortex abnormality we found in our borderline subjects may be considered a neurodevelopmental deficit of the right hemisphere. Individuals with BPD and PTSD were reported to have a considerably increased prevalence of subtle neurologic impairment (Gurvits et al 2000; van Reekum et al 1996). Weintraub and Mesulam (1983) described a behavioral syndrome characterized by emotional and interpersonal difficulties, shyness, visuospatial disturbances, and inadequate paralinguistic communicative abilities related to neurologic signs of right hemisphere dysfunction. The BPD subjects in our study displayed marked deficits in the domain of attention, visual memory, and visuospatial cognition (Table 3). Similar neuropsychologic findings have been obtained earlier in individuals with BPD (O’Leary et al 1991). www.elsevier.com/locate/biopsych

180 BIOL PSYCHIATRY 2005;57:173–182 Currently, the role of the right parietal cortex is modeled as an epicenter for the mental representation and attentional targeting of salient extrapersonal events (Mesulam 1999). Visuospatial and attentional functions are most likely compromised after lesions of the right posterior parietal cortex (Mesulam 1998). Accordingly, the attentional performance of our control subjects was significantly related to parietal cortex size (Table 3), which may serve as a kind of validation of the parietal cortex measure used in this study. Attentional deficits as well as visuospatial and memory deficits of borderline subjects, however, were related to small right hippocampal size instead (Table 3). These results support earlier studies demonstrating a strong relationship between hippocampal size and cognitive functioning of individuals who had been exposed to traumatic stress (Bremner et al 1995a, 1997; Gurvits et al 1996). The mechanism for developing small right parietal cortices could also be sought in the traumatic life history of our borderline subjects. It could be speculated that early and repeated interpersonal traumatic events lead to a pervasive pattern of detachment from social relationships and thus to an understimulation of the right parietal cortex. Recent evidence suggests that maturation of parietal and frontal cortices continues into adulthood (Sowell et al 2003), rendering them likely candidates for environmental plasticity. An exploratory analysis of our data, however, did not reveal any relationships among the severity of traumatic experiences (TAQ), trauma-related clinical symptoms (IES-R), or dissociative symptoms (DES) and right parietal cortex volume or parietal cortex asymmetry, respectively. Hippocampal Size Reduction in BPD Our results complement earlier findings on reduced hippocampal size in adult survivors of childhood abuse with BPD (Driessen et al 2000; Rüsch et al 2003; Schmahl et al 2003a; Tebartz van Elst et al 2003) or without BPD (Bremner et al 1997, 2003; Stein et al 1997; Villarreal et al 2002; Vythilingam et al 2002). We found smaller hippocampal volumes in borderline subjects with stronger trauma-related clinical symptoms. Similar relationships between PTSD symptom severity and reduced hippocampal size were reported for adult survivors of childhood abuse (Bremner et al 1997, 2003; Villarreal et al 2002), combat veterans (Gilbertson et al 2002; Gurvits et al 1996), and adult burn survivors (Winter and Irle 2004). Our results might argue for a detrimental influence of prolonged trauma-related distress on hippocampal size. Preclinical evidence suggests that stress-related hyperglutamatergic states may induce hippocampal degeneration in individuals with traumatic exposure (Chambers et al 1999). Evidence also suggests that small hippocampal size is a risk factor for developing PTSD in Vietnam combat veterans (Gilbertson et al 2002) and that lower precombat intelligence predicts exposure to more severe combat (Macklin et al 1998). Our borderline subjects with the additional diagnosis of PTSD presented with more neuropsychologic deficits and smaller brain size than subjects without PTSD and reported more childhood sexual abuse. Early childhood sexual abuse in women was shown to be a significant predictor of both BPD and PTSD (McLean and Gallop 2003). Others have reported neuropsychologic deficits in individuals with childhood abuse (Bremner et al 1995b). Future studies are needed to elucidate the specific role of early sexual abuse on brain size and development. Methodologic Considerations BPD subjects with the diagnosis of any current or lifetime psychotic disorder were rigorously excluded. Further exclusion www.elsevier.com/locate/biopsych

E. Irle et al criteria were a history of neurologic disorders, electroencephalograph abnormalities, or hyperintensive MRI signals. As a consequence, the spectrum and variability of psychotic symptoms and DSM-IV cluster A personality traits and the severity of brain abnormalities may have been reduced in our sample. Nevertheless, we found a significant right parietal cortex size reduction (Figure 2) and a relationship between parietal cortex asymmetry and psychotic symptoms and schizoid personality traits (Figures 5–7). Our BPD sample presented with a high DSM-IV Axis I comorbidity. Therefore, the findings of our study must be considered with caution. Concurrent Axis I disorders are found in the vast majority of individuals with BPD (Zanarini et al 1998), leading to the conclusion that any BPD sample being limited to subjects with the sole diagnosis of BPD cannot be considered representative (Skodol et al 2002). Our results do not give evidence for an influence of comorbid disorders on the size of the parietal cortex or the hippocampus of our BPD subjects. We cannot completely rule out that comorbid depressive disorder had an influence on our findings, however. Twenty-six of our BPD subjects met criteria for current or lifetime major depression, thus preventing a more rigorous test of subgroups (i.e., subjects with and without major depression). Our findings may not be valid for men with BPD. Previous research has demonstrated considerable gender differences in the temporoparietal cortex asymmetry of normal individuals and individuals with schizophrenia or schizotypal personality disorder (Dickey et al 2002, 2003; Frederikse et al 2000; Niznikiewicz et al 2000; Shapleske et al 1999). Conclusions and Directions for Future Research Our results demonstrate reduced size and abnormal asymmetry of the parietal cortex in BPD. Reduced right parietal cortex size in BPD may reflect a neurodevelopmental deficit of the right hemisphere. It might be speculated that a reduced leftward asymmetry of temporoparietal cortices increases the likeliness for developing psychotic symptoms, whereas increased leftward parietal cortex asymmetry may protect individuals with BPD and PTSD from developing more pronounced and disabling psychotic symptoms. Studies are currently being done in our department to determine whether specific posterior cortical areas, especially the precuneus and posterior cingulate cortex, have a reduced size in subjects with BPD. A current positron emission tomography study (Lange et al, unpublished data) has revealed reduced resting glucose metabolism in the precuneus and posterior cingulate cortex of a subgroup of 17 borderline subjects who are also included in this report. This research was supported by Grant No. IR 15/8 (to EI) from the Deutsche Forschungsgemeinschaft. We thank the subjects who participated in this study and E. Koscholke, N. Jacobs, V. Prischmann, and C. Galwas for assistance in the subjects’ assessments. American Psychiatric Association (1994): Diagnostic and Statistical Manual of Mental Disorders, 4th ed. Washington, DC: American Psychiatric Press. Barta PE, Pearlson GD, Powers RE (1990): Auditory hallucinations and smaller superior temporal gyrus volume in schizophrenia. Am J Psychiatry 147: 1457–1462. Beck AT, Ward CH, Mendelson M, Mock J, Erbaugh J (1960): An inventory for measuring depression. Arch Gen Psychiatry 4:561–571. Bernstein EM, Putnam FW (1986): Development, reliability, and validity of a dissociation scale. J Nerv Ment Dis 174:727–735.

E. Irle et al Bremner JD, Randall PR, Scott TM, Bronen RA, Seibyl JP, Southwick SM, et al (1995a): MRI-based measurement of hippocampal volume in patients with combat-related posttraumatic stress disorder. Am J Psychiatry 152:973–981. Bremner JD, Randall P, Scott TM, Capelli S, Delaney R, McCarthy G, et al (1995b): Deficits in short-term memory in adult survivors of childhood abuse. Psychiatry Res 59:97–107. Bremner JD, Randall PR, Vermetten E, Staib L, Bronen RA, Mazure C, et al (1997): Magnetic resonance imaging– based measurement of hippocampal volume in posttraumatic stress disorder related to childhood physical and sexual abuse—a preliminary report. Biol Psychiatry 41:23– 32. Bremner JD, Vythilingam M, Vermetten E, Southwick SM, McGlashan T, Nazeer A, et al (2003): MRI and PET study of deficits in hippocampal structure and function in women with childhood sexual abuse and posttraumatic stress disorder. Am J Psychiatry 160:924 –932. Breslau N, Davis GC, Andreski P, Peterson E (1991): Traumatic events and posttraumatic stress disorder in an urban population of young adults. Arch Gen Psychiatry 48:216 –222. Brunner E, Domhof S, Langer F (2002): Nonparametric Analysis of Longitudinal Data in Factorial Experiments. New York: Wiley. Butler RW, Mueser KT, Sprock J, Braff DL (1996): Positive symptoms of psychosis in posttraumatic stress disorder. Biol Psychiatry 39:839 – 844. Chambers RA, Bremner JD, Moghaddam B, Southwick SM, Charney DS, Krystal JH (1999): Glutamate and post-traumatic stress disorder: Toward a psychobiology of dissociation. Semin Clin Neuropsychiatry 4:274 –281. David D, Kutcher GS, Jackson EI, Melman TA (1999): Psychotic symptoms in combat-related posttraumatic stress disorder. J Clin Psychiatry 60:29 –32. De la Fuente JM, Goldman S, Stanus E, Vizuete C, Morlan I, Bobes J, et al (1997): Brain glucose metabolism in borderline personality disorder. J Pychiatr Res 31:531–541. Dickey CC, McCarley RW, Voglmaier MM, Frumin M, Niznikiewicz M, Hirayasu Y, et al (2002): Smaller left Heschl’s gyrus volume in patients with schizotypal personality disorder. Am J Psychiatry 159:1521–1527. Dickey CC, McCarley RW, Voglmaier MM, Niznikiewicz MA, Seidman LJ, Demeo S, et al (2003): An MRI study of superior temporal gyrus volume in women with schizotypal personality disorder. Am J Psychiatry 160:2198 – 2201. Donegan NH, Sanislow CA, Blumberg HP, Fulbright RK, Lacadie C, Skudlarski P, et al (2003): Amygdala hyperreactivity in borderline personality disorder: Implications for emotional dysregulation. Biol Psychiatry 54:1284 – 1293. Driessen M, Herrmann J, Stahl K, Zwaan M, Meier S, Hill A, et al (2000): Magnetic resonance imaging volumes of the hippocampus and the amygdala in women with borderline personality disorder and early traumatization. Arch Gen Psychiatry 57:1115–1122. Duvernoy HM (1998): The Human Hippocampus. Functional Anatomy, Vascularization, and Serial Sections with MRI. Berlin: Springer. Feuerlein W, Küfner H, Ringer C, Antons K (1979): Münchner AlkoholismusTest (MALT). Weinheim FRG: Beltz. First MB, Gibbon M, Spitzer RL, Williams JBW, Benjamin L (1997): Structured Clinical Interview for DSM-IV Axis II Personality Disorders (SCID-II). Washington, DC: American Psychiatric Press. First MB, Spitzer RL, Gibbon M, Williams JBW (1995): Structured Clinical Interview for DSM-IV (SCID-I). New York: Biometrics Research Department, New York State Psychiatric Institute. Frederikse ME, Lu A, Aylward E, Barta P, Sharma T, Pearlson G (2000): Sex differences in inferior parietal lobule volume in schizophrenia. Am J Psychiatry 157:422– 427. Freyberger HJ, Spitzer C, Stieglitz R-D (1999): Fragebogen zu Dissoziativen Symptomen. Bern, Switzerland: Huber. Gaser C, Nenadic I, Volz HP, Büchel C, Sauer H (2004): Neuroanatomy of “hearing voices”: A frontotemporal brain structural abnormality associated with auditory hallucinations in schizophrenia. Cereb Cortex 14:91– 96. Gast U, Oswald T, Zündorf F, Hofmann A (2000): Strukturiertes Klinisches Interview für DSM-IV. Dissoziative Störungen. Göttingen, Germany: Hogrefe. Gilbertson MW, Shenton ME, Ciszewski A, Kasai K, Lasko NB, Orr SP, et al (2002): Smaller hippocampal volume predicts pathological vulnerability to psychological trauma. Nat Neurosci 5:1242–1247. Gunderson JG, Kolb JE (1981): The diagnostic interview for borderline patients. Am J Psychiatry 132:896 –903.

BIOL PSYCHIATRY 2005;57:173–182 181 Gurvits TV, Gilbertson MW, Lasko NB, Tarhan AS, Simeon D, Macklin ML, et al (2000): Neurological soft signs in chronic posttraumatic stress disorder. Arch Gen Psychiatry 57:181–186. Gurvits TV, Shenton ME, Hokama H, Ohta H, Lasko NB, Gilbertson MW, et al (1996): Magnetic resonance imaging study of hippocampal volume in chronic, combat-related posttraumatic stress disorder. Biol Psychiatry 40:1091–1099. Hamner MB (1997): Psychotic features and combat-associated PTSD. Depress Anxiety 5:34 –38. Hamner MB, Frueh BC, Ulmer HG, Arana GW (1999): Psychotic features and illness severity in combat veterans with chronic posttraumatic stress disorder. Biol Psychiatry 45:846 – 852. Hautzinger M, Bailer M, Worall H, Keller F (1995): Beck Depressions Inventar (BDI). Bern, Switzerland: Huber. Herman JL, Perry JC, van der Kolk BA (1989): Childhood trauma in borderline personality disorder. Am J Psychiatry 146:490 – 495. Herpetz SC, Dietrich TM, Wenning B, Krings T, Erberich SG, Willmes K, et al (2001): Evidence of abnormal amygdala functioning in borderline personality disorder: A functional MRI study. Biol Psychiatry 50:292–298. Hoffman RE, Hawkins KA, Gueorguieva R, Boutros NN, Rachid F, Carroll K, et al (2003): Transcranial magnetic stimulation of left temporoparietal cortex and medication-resistant auditory hallucinations. Arch Gen Psychiatry 60:49 –56. Juengling FD, Schmahl C, He␤linger B, Ebert D, Bremner JD, Gostomzyk J, et al (2003): Positron emission tomography in female patients with borderline personality disorder. J Psychiatr Res 37:109 –115. Kernberg OF (1967): Borderline personality organization. Am J Psychoanal Assoc 15:641– 685. Kernberg OF, Dulz B, Sachsse U (2000): Handbuch der Borderline-Störungen. Stuttgart: Schattauer. Krystal JH, Karper LP, Seibyl JP, Freeman GK, Delaney R, Bremner JD, et al (1994): Subanesthetic effects of the noncompetitive NMDA antagonist, ketamine, in humans. Arch Gen Psychiatry 51:199 –214. Kwon JS, McCarley RW, Hirayasu Y, Anderson JE, Fischer IA, Kikinis R, et al (1999): Left planum temporale volume reduction in schizophrenia. Arch Gen Psychiatry 56:142–148. Macklin ML, Metzger LJ, Litz BT, McNally RJ, Lasko NB, Orr SP, et al (1998): Lower precombat intelligence is a risk factor for posttraumatic stress disorder. J Consult Clin Psychol 66:323–326. Maercker A, Schützwohl M (1998): Erfassung von psychischen Belastungsfolgen: Die Impact of Events Skala—revidierte Version (IES-R). Diagnostica 44:130 –141. McLean LM, Gallop R (2003): Implications of childhood sexual abuse for adult borderline personality disorder and complex posttraumatic stress disorder. Am J Psychiatry 160:369 –371. Mesulam MM (1998): From sensation to cognition. Brain 121:1013–1052. Mesulam MM (1999): Spatial attention and neglect: Parietal, frontal and cingulate contributions to the mental representation and attentional targeting of salient extrapersonal events. Phil Trans R Soc Lond B 354: 1325–1346. Niznikiewicz M, Donnino R, McCarley RW, Nestor PG, Iosifescu DV, O’Donnell B, et al (2000): Abnormal angular gyrus asymmetry in schizophrenia. Am J Psychiatry 157:428 – 437. O’Leary KM, Brouwers P, Gardner DL, Cowdry RW (1991): Neuropsychological testing of patients with borderline personality disorder. Am J Psychiatry 148:106 –111. Olney JW, Farber NB (1995): Glutamate receptor dysfunction and schizophrenia. Arch Gen Psychiatry 52:998 –1007. Pearlson GD, Petty RG, Ross CA, Tien AY (1996): Schizophrenia: A disease of heteromodal association cortex? Neuropsychopharmacology 14:1–17. Pruessner JC, Li LM, Serles W, Pruessner M, Collins DL, Kabani N, et al (2000): Volumetry of hippocampus and amygdala with high-resolution MRI and three-dimensional analysis software: Minimizing the discrepancies between laboratories. Cereb Cortex 10:433– 442. Pütterich H (1990): Diagnostisches Interview für das Borderline-Syndrom. Weinheim, Germany: Beltz. Rademacher J, Galaburda AM, Kennedy DN, Filipek PA, Caviness VS Jr (1992): Human cerebral cortex: Location, parcellation, and morphometry with magnetic resonance imaging. J Cog Neurosci 4:352–374. Rüsch N, Tebartz van Elst T, Ludaescher P, Wilke M, Huppertz HJ, Thiel T, et al (2003): A voxel-based morphometric MRI study in female patients with borderline personality disorder. NeuroImage 20:385–392.

www.elsevier.com/locate/biopsych

182 BIOL PSYCHIATRY 2005;57:173–182 Sautter FJ, Bissette G, Wiley J, Manguno-Mire G, Schoenbachler B, Myers L, et al (2003): Corticotropin-releasing factor in posttraumatic stress disorder (PTSD) with secondary psychotic symptoms, nonpsychotic PTSD and healthy control subjects. Biol Psychiatry 54:1382–1388. Sautter FJ, Cornwell J, Johnson JJ, Wiley J, Faraone SV (2002): Family history study of posttraumatic stress disorder with secondary psychotic symptoms. Am J Psychiatry 159:1775–1777. Schmahl CG, Elzinga BM, Vermetten E, Sanislow C, McGlashan TH, Bremner JD (2003b): Neural correlates of memories of abandonment in women with and without borderline personality disorder. Biol Psychiatry 54:142–151. Schmahl CG, Vermetten E, Elzinga BM, Bremner JD (2003a): Magnetic resonance imaging of hippocampal and amygdala volume in women with childhood abuse and borderline personality disorder. Psychiatry Res Neuroimaging 122:193–198. Shapleske J, Rossell SL, Woodruff PWR, David AS (1999): The planum temporale: A systematic, quantitative review of its structural, functional and clinical significance. Brain Res Rev 29:26 – 49. Shenton ME, Dickey CC, Frumin M, McCarley RW (2001): A review of MRI findings in schizophrenia. Schizophr Res 49:1–52. Shenton ME, Kikinis R, Jolesz FA, Pollak SD, LeMay M, Wible CG, et al (1992): Left temporal lobe abnormalities in schizophrenia and thought disorder: A quantitative MRI study. New Engl J Med 27:604 – 612. Skodol AE, Gunderson JG, Pfohl B, Widiger TA, Livesley WJ, Siever LJ (2002): The borderline diagnosis I: Psychopathology, comorbidity, and personality structure. Biol Psychiatry 51:936 –950. Soloff PH, Meltzer CC, Becker C, Greer PJ, Kelly TM, Constantine D (2003): Impulsivity and prefrontal hypometabolism in borderline personality disorder. Psychiatry Res 123:153–163. Soloff PH, Meltzer CC, Greer PJ, Constantine D, Kelly TM (2000): A fenfluramine-activated FDG-PET study of borderline personality disorder. Biol Psychiatry 47:540 –547. Sommer I, Aleman A, Ramsey N, Bouma A, Kahn R (2001): Handedness, language lateralisation and anatomical asymmetry in schizophrenia. Brit J Psychiatry 178:344 –351. Sowell ER, Peterson BS, Thompson PM, Welcome SE, Henkenius AL, Toga AW (2003): Mapping cortical change across the human life span. Nat Neurosci 6:309 –315. Stein MB, Koverola C, Hanna C, Torchia MG, McClarty B (1997): Hippocampal volume in women victimized by childhood sexual abuse. Psychol Med 27:951–959.

www.elsevier.com/locate/biopsych

E. Irle et al Steinberg M (1994): Structured Clinical Interview for DSM-IV Dissociative Disorders—Revised (SCID-D). Washington, DC: American Psychiatric Press. Tebartz Van Elst L, Hesslinger B, Thiel T, Geiger E, Haegele K, Lemieux L, et al (2003): Frontolimbic brain abnormalities in patients with borderline personality disorder: A volumetric magnetic resonance imaging study. Biol Psychiatry 54:163–171. Tewes U (1991): HAWIE-R. Hamburg-Wechsler-Intelligenztest für Erwachsene. Bern, Switzerland: Huber. [revised edition] van Reekum R, Links PS, Finlayson AJ, Boyle M, Boiago I, Ostrander AL, et al (1996): Repeat neurobehavioral study of borderline personality disorder. J Psychiatry Neurosci 21:13–20. Villarreal G, Hamilton DA, Petropoulos H, Driscoll I, Rowland LM, Griego JA, et al (2002): Reduced hippocampal volume and total white matter volume in posttraumatic stress disorder. Biol Psychiatry 52:119 –125. Vythilingam M, Heim C, Newport J, Miller AH, Anderson E, Bronen R, et al (2002): Childhood trauma associated with smaller hippocampal volume in women with major depression. Am J Psychiatry 159:2072–2080. Wechsler D (1981): Wechsler Adult Intelligence Scale—Revised (WAIS-R). New York: Psychological Corporation. Wechsler D (1987): Wechsler Memory Scale—Revised (WMS-R). San Antonio, TX: Psychological Corporation. Weintraub S, Mesulam MM (1983): Developmental learning disabilities of the right hemisphere. Arch Neurol 40:463– 468. Weiss DS, Marmar CR (1996): The impact of event scale-revised. In: Wilson JP, Keane TM, editors. Assessing Psychological Trauma and PTSD. New York: Guilford Press, 399 – 441. White CN, Gunderson JG, Zanarini MC, Hudson JI (2003): Family studies of borderline personality disorder: a review. Harv Rev Psychiatry 11:8 –19. Winter H, Irle E (2004): Reduced hippocampal volume in adult burn survivors with and without posttraumatic stress disorder. Am J Psychiatry 161: 2197-2200. Wittchen HU, Zaudig M, Frydrich T (1997): Strukturiertes Klinisches Interview für DSM-IV. Achse I und II. Göttingen, Germany: Hogrefe. Zanarini MC, Frankenburg FR, Dubo ED, Sickel AE, Trikha A, Levin A, et al (1998): Axis I comorbidity of borderline personality disorder. Am J Psychiatry 155:1733–1739. Zimmermann P, Fimm B (1993): Testbatterie zur Aufmerksamkeitsprüfung (TAP), version 1.02. Freiburg, Germany: Psytest.