Reduced cortical and subcortical volumes in female adolescents with borderline personality disorder

Psychiatry Research: Neuroimaging 221 (2014) 179–186

Contents lists available at ScienceDirect

Psychiatry Research: Neuroimaging journal homepage: www.elsevier.com/locate/psychresns

Reduced cortical and subcortical volumes in female adolescents with borderline personality disorder Julia Richter a,b, Romuald Brunner a, Peter Parzer a, Franz Resch a, Bram Stieltjes b, Romy Henze a,b,n a Section Disorders of Personality Development, Department of Child and Adolescent Psychiatry, Center for Psychosocial Medicine, University Hospital of Heidelberg, Heidelberg, Germany b Section Quantitative Imaging-Based Disease Characterization, Department of Radiology, German Cancer Research Center, Heidelberg, Germany

art ic l e i nf o

a b s t r a c t

Article history: Received 17 July 2013 Received in revised form 13 January 2014 Accepted 17 January 2014 Available online 23 January 2014

Volumetric alterations in limbic structures have been detected in adults, but not in adolescents with borderline personality disorder (BPD). We examined adolescents in the early stages of BPD to provide a unique opportunity to investigate which parts of the brain are initially affected by the disorder before confounding factors such as long-term medication or chronicity can mask them. A group of 60 righthanded female adolescents between 14 and 18 years of age (20 patients with BPD, 20 clinical controls, and 20 healthy controls) underwent magnetic resonance imaging (MRI). Focus was on the examination of hippocampal and amygdalar volume differences. Furthermore, a cortical thickness analysis was conducted. FreeSurfer software detected significant group differences in the right and left hippocampus and in the right amygdala. Additionally, significant volume reductions in frontal (right middle frontal gyrus, orbital part of the inferior frontal gyrus bilaterally), and parietal regions (superior parietal gyrus bilaterally) were found in adolescents with BPD compared with controls. No group differences in cortical thickness were revealed. & 2014 Elsevier Ireland Ltd. All rights reserved.

Keywords: FreeSurfer Borderline personality disorder Parietal cortex Amygdala Hippocampus Cortical thickness

1. Introduction Subjects who are affected by borderline personality disorder (BPD) suffer from mood instability and impulsiveness, show dysregulated behaviors such as non-suicidal self-injury, or have a history of multiple suicide attempts (Selby et al., 2009). They have difficulties in maintaining interpersonal relationships or trusting other people, and they may also experience cognitive symptoms such as paranoia or severe dissociative symptoms. The prevalence of BPD in community samples is estimated to be approximately 1% (Paris, 2005) and rises in psychiatric settings to 10% in outpatients and 25% in inpatients (Leichsenring et al., 2011). Several studies demonstrated that BPD is related to neuroanatomical changes (Schmahl and Bremner, 2006). However, there are differences between adults and adolescents with BPD. 1.1. Neuroimaging findings in adults with BPD Adults with BPD exhibit structural and functional alterations in limbic structures such as the amygdala and the hippocampus, but n Corresponding author at: Section Disorders of Personality Development, Department of Child and Adolescent Psychiatry, Center for Psychosocial Medicine, University Hospital of Heidelberg, Heidelberg, Germany. Tel.: þ 49 6221 5636002. E-mail address: [email protected] (R. Henze).

0925-4927/$ - see front matter & 2014 Elsevier Ireland Ltd. All rights reserved. http://dx.doi.org/10.1016/j.pscychresns.2014.01.006

also in prefrontal regions, in the cingulate cortex and in the parietal lobule in comparison to healthy controls (HC) (Irle et al., 2005, 2007; Lis et al., 2007). Limbic structures, the prefrontal regions, and the anterior cingulate cortex (ACC) have functions that are related to the psychopathology of BPD, such as impulsivity, emotional instability, and impulsive aggression (Tebartz van Elst et al., 2003). With regard to limbic structures, studies reported alterations in the hippocampus and the amygdala in adults with BPD (Schmahl and Bremner, 2006; Nunes et al., 2009). A well-replicated result is the finding of a smaller volume in the hippocampus bilaterally in adult patients with BPD in comparison to HC (Driessen et al., 2000; Schmahl et al., 2003; Tebartz van Elst et al., 2003; Nunes et al., 2009; Weniger et al., 2009). Although the results concerning the amygdala are not as conclusive as the results regarding the hippocampus in adult subjects with BPD, there are several studies indicating alterations in the amygdala in adult samples with BPD. However, the direction of the alteration is not yet established. Some studies and a meta-analysis revealed a volumetric reduction of the amygdala bilaterally (Driessen et al., 2000; Schmahl et al., 2003; Tebartz van Elst et al., 2003; Nunes et al., 2009; Weniger et al., 2009), whereas another study only detected a grey matter loss in the left amygdala (Rüsch et al., 2003). One study even found increased grey matter concentrations in the amygdala bilaterally (Minzenberg et al., 2008). A bilateral enlargement of the amygdala was detected in patients with BPD and comorbid depression (Zetzsche et al., 2006).

180

J. Richter et al. / Psychiatry Research: Neuroimaging 221 (2014) 179–186

Still under debate is the impact of comorbid post-traumatic stress disorder (PTSD) on limbic grey matter volumes in patients with BPD. Three meta-analyses and a very recently published study examined the influence of PTSD on amygdalar and hippocampal volumes in patients with BPD, but they yielded conflicting findings (no effect of comorbid PTSD: Niedtfeld et al., 2013; Ruocco et al., 2012; decreased amygdalar volume in BPD without comorbid PTSD: de-Almeida et al., 2012; decreased hippocampal volumes in BPD with comorbid PTSD: Rodrigues et al., 2011). With regard to frontal brain structures and the ACC, a smaller frontal lobe (Lyoo et al., 1998), significant volume reductions of the left orbitofrontal cortex (OFC) and the right ACC (Tebartz van Elst et al., 2003), reduced grey and increased white matter volume of the ACC and the posterior cingulate cortex (Hazlett et al., 2005), and reduced grey matter concentrations in the left rostral/subgenual ACC (Minzenberg et al., 2008) were found in adults with BPD when compared with HC. In parietal areas, grey matter volume decreases were detected in the right parietal cortex (Irle et al., 2005) and in the right superior parietal cortex in female adults with BPD (Irle et al., 2007), as well as in the left superior parietal cortex and the right inferior parietal lobe in men with BPD when compared with HC (Völlm et al., 2009). Concerning white matter fiber tracks, findings include a decreased white matter microstructural integrity in the inferior frontal brain region (Grant, et al., 2007), a decreased fractional anisotropy (FA) in the genu and rostral areas of the corpus callosum and of the prefrontal white matter fasciculus bilaterally in adults with BPD (Carrasco et al., 2012), and a reduced interhemispheric connectivity of the left and right dorsal ACCs in patients with BPD and comorbid ADHD compared with HC (Rüsch et al., 2010). 1.2. Neuroimaging findings in adolescents with BPD Neuroanatomical changes in adult subjects with BPD are probably not only caused by neuroanatomical correlates of the disorder itself, but also by confounding factors such as long-term medication or the process of chronicity. Neuroimaging studies investigating adolescent subjects with BPD are sparse, but they shed light on neuroanatomical changes caused by BPD directly, since adolescents have in most cases not received long-term medication and effects of chronicity are minimized. In contrast to adult samples with BPD, no volumetric differences in limbic structures have been revealed in adolescents. Only a higher grey matter volume in the left OFC than in the right OFC was reported in adolescents with BPD, indicating a reversal of the normal right-left OFC symmetry in HC (Chanen et al., 2008). Other studies report a reduced grey matter volume of the left ACC (Whittle et al., 2009) and of the ACC bilaterally (Goodman et al., 2011) as well as a shortened adhesio interthalamica (Takahashi et al., 2009) in subjects with BPD when compared with HC. In a very recent diffusion tensor imaging study, reduce FA in the inferior longitudinal fasciculus, in the uncinate, and in the occipitofrontal fasciculi was found in adolescents with BPD compared with HC (New et al., 2013). In our own voxel-based morphometry (VBM) study using statistical parametric mapping (SPM5) (Ashburner and Friston, 2005, 2000) as the method of analysis, we detected a volumetric reduction in the left OFC and the dorsolateral prefrontal cortex (DLPFC) bilaterally in adolescent females with BPD in comparison to HC, but no changes in any limbic structure (Brunner et al., 2010). Analyzing the same data set with regard to differences in white matter fiber tracks, we found white matter changes in a part of the limbic system, namely the fornix, in inferior frontal brain areas as well as more widespread changes in areas related to the heteromodal association cortex (HASC) (Buchanan et al., 2004; Ross and

Pearlson, 1996) in the sample of BPD subjects, indicating an involvement of brain areas related to emotion regulation and emotion cognition (Maier-Hein et al., 2014). Thus, our morphometric study points to alterations in BPD restricted only to prefrontal brain areas, whereas our study analyzing white matter fiber tracks argues for a more widespread involvement of different brain structures in adolescents with BPD, including the limbic system. 1.3. Aims and hypotheses In our earlier VBM study using SPM5 (Ashburner and Friston, 2005, 2000), we did not only expect volumetric changes in frontal areas, but also in limbic structures, in the BPD group when compared with clinical controls (CC) and HC (Brunner et al., 2010). However, we could not detect alterations in any limbic structure using SPM5. Analyzing the same data set with regard to differences of white matter fiber tracts between the groups, we detected a more widespread involvement of several brain areas, also including a structure of the limbic system (Maier-Hein et al., 2014). This divergence of the results of our studies, although obtained in the same data set, led us to reanalyze the volumes between the groups using another technique than SPM5. We decided to reanalyze the data using FreeSurfer software (Dale et al., 1999; Fischl et al., 1999) as FreeSurfer is claimed to be an excellent method to analyze subcortical structures. Two earlier studies that compared different analytic methods in adult samples concluded that FreeSurfer displayed the best volumetric results with regard to limbic structures (Tae et al., 2008; Dewey et al., 2010). Assuming that FreeSurfer would show the same high-quality standards when analyzing limbic structures in adolescent samples, we deemed it a suitable method for a reanalysis of the data set. We expected to find alterations in limbic structures, especially the amygdala and the hippocampus, in adolescent girls with BPD when compared with CC and HC. Furthermore, we expected volumetric alterations in frontal areas in the BPD group as already shown by our own VBM study (Brunner et al., 2010) and a study by Chanen et al. (2008). Based on the results of other studies using samples with subjects suffering from BPD, we expected volumetric decreases in the ACC and in the parietal lobe (Irle et al., 2007; Chanen et al., 2008; Takahashi et al., 2009; Brunner et al., 2010; Goodman et al., 2011). Additionally, we used FreeSurfer to examine whether adolescents with BPD exhibit alterations in cortical thickness in neuroanatomical structures that showed volumetric differences in previous studies (Chanen et al., 2008; Takahashi et al., 2009; Whittle et al., 2009; Brunner et al., 2010; Goodman et al., 2011). Cortical thickness is, together with the cortical surface area, a determinant of the size of grey matter volume (Hutton et al., 2008; Panizzon et al., 2009; Winkler et al., 2010). Thus, an examination of cortical thickness may give a deeper insight into the underlying mechanisms responsible for volumetric changes of neuroanatomical structures in adolescents with BPD. Due to the results of previous studies, we expected reduced cortical thickness in the BPD group in frontal and cingulate brain areas (Chanen et al., 2008; Takahashi et al., 2009; Whittle et al., 2009; Brunner et al., 2010; Goodman et al., 2011).

2. Methods 2.1. Subjects The sample consisted of 60 right-handed adolescent girls between 14 and 18 years of age. Patients with a lifetime diagnosis of schizophrenia, schizoaffective disorder, bipolar disorder, pervasive developmental disorder, alcohol or drug dependence, or neurological disease were excluded as well as girls with a body

J. Richter et al. / Psychiatry Research: Neuroimaging 221 (2014) 179–186 mass index o 16 or with an IQ r 85. The participants were divided into three groups: a group of patients with BPD diagnosed using the Diagnostic and Statistical Manual of Mental Disorders (DSM) IV (American Psychiatric Association, 2000); a group of CC with other psychiatric disorders (not fulfilling more than one criterion of the BPD diagnosis); and a group of HC who had no current psychiatric disorder, had never received a psychiatric diagnosis before, and who had never received any psychiatric or psychological treatment. Each group comprised 20 participants. Patients were recruited from the Department of Child and Adolescent Psychiatry at the University Hospital of Heidelberg and were informed about the study by their physician. The group of HC was acquired through advertisements in public schools. Subjects were included after an assessment of handedness and a confirmation of diagnosis for either the group of patients with BPD or the group of CC. All participants were assessed in a structured clinical interview that checked for comorbid psychiatric disorders and the presence or absence of a psychiatric disorder in HC. Furthermore, HC and CC underwent the BPD section of a structural clinical interview for personality disorders. All included patients were matched for age and school type. The study was approved by the Ethics committee of the Faculty of Medicine, University of Heidelberg. Both participants and their legal guardians gave written informed consent after the study and its procedure was explained to them. 2.2. Outcome measures A diagnosis of BPD was obtained using the German version (Fydrich et al., 1997) of the BPD section of the structural clinical interview for DSM IV Axis II Personality Disorders (SCID-II) (First et al., 1997). To assess comorbid psychiatric disorders and to confirm the lack of psychiatric disorders in HC, the German version (Delmo et al., 2000) of the Schedule for Affective Disorders and Schizophrenia for School-Age Children-Present (K-SADS-P) was used (Kaufman et al., 1997). Interviews were conducted by trained graduate-level clinicians, and uncertain ratings were discussed with research staff. The interview of the BPD section of the SCID-II was tape-recorded so that it could be reexamined later. A sample of 30 subjects was selected at random and the BPD diagnosis was confirmed by a second rater with a satisfactory interrater reliability (Cohen's Kappa ¼ 0.94). Handedness was measured using the Edinburgh Handedness Inventory (Oldfield, 1971) and IQ using the German version (Dahl, 1986) of the Wechsler Abbreviated Intelligence Scale (Wechsler, 1999). Several other psychometric measurements such as psychosocial functioning in general (Children's Global Assessment Scale (C-GAS)) (Delmo et al., 2000; Shaffer et al., 1983) and the occurrence of other disorders’ symptomatology were also obtained (Beck, 1995; Nader et al., 1996) A more detailed overview of the different variables is shown in Table 1.

181

2.3. MRI scanning The structural MRI datasets were acquired with a three-dimensional sagittal isotropic Magnetization Prepared Rapid Acquisition Gradient Echo (MPRAGE) sequence and a 12-channel standard head coil using a 3 T scanner (Tim Trio, Siemens, Erlangen, Germany). The parameters applied were as follows: flip angle 91, repetition time 2300 ms, echo time 2.98 s, field of view 256 mm, matrix size 256  256 pixels, slice thickness 1 mm, 160 slices. MR images were reviewed by an experienced neuroradiologist.

2.4. Data processing and analysis For the investigation of the volumetric data, FreeSurfer automatically classifies subcortical and cortical structures (Caviness et al., 1989; Fischl et al., 2002). The segmentation of MR images for volumetric analysis includes computing an optimal linear transformation and a non-linear morphing. The segmentation process uses three pieces of information: the prior probability of a given tissue class occurring at a specific atlas location, the likelihood of the image intensity given that tissue class, and the probability of the local spatial configuration of labels for that tissue class (Walhovd et al., 2005). For the analysis, the Destrieux atlas was used (Fischl et al., 2004). The data were analyzed using analysis of co-variance (ANCOVA) for continuous variables with intracranial volume (ICV) as covariate. For pairwise post-hoc comparisons of the groups, F-values were corrected using Sidak's method to account for multiple testing. The analysis was done using the statistical software program Stata, version 12 (StataCorp, 2011). Cortical thickness was also analyzed using FreeSurfer (available under http:// surfer.nmr.mgh.harvard.edu/). Processing in FreeSurfer includes correction for magnetic field inhomogeneities and skull stripping to remove non-brain tissue. White matter is segmented, a mesh representation around the white matter surface is generated, and the surface is smoothed (Clarkson et al., 2011). Topological defects are corrected automatically. After the initial surface model is constructed, a second smoothing procedure is applied to obtain a representation of the grey/white matter boundary which is expanded to the pial surface afterwards. To get a measurement of cortical thickness and to compare it between subjects, the data is registered to an average spherical surface, and the shortest distance between a given point on the grey/white matter boundary and pial surface and vice versa is calculated and averaged (Fischl et al., 1999; Fischl and Dale, 2000). The cortex of every individual subject was visually inspected and manually corrected if necessary. A cross-subject general linear model (GLM) was used to test group-wise differences in cortical thickness between girls with BPD, CC and HC. Individual subject measurements were smoothed using a full width at half maximum

Table 1 Means, standard deviations, P-values and post-hoc comparisons of demographic and psychometric measures in subjects with borderline personality disorder (n¼ 20), clinical controls (n ¼20) and healthy controls (n¼ 20). Measures

BPD

CC

HC

F(2,57)

P

Age, mean (SD), y IQ, mean (SD) School type, no. (%) Gymnasium Realschule Hauptschule Clinical setting, no. (%) Inpatients Day clinic Outpatients BDIa, mean (SD) BAIa, mean (SD) A-DESa, mean (SD) BISa, mean (SD) ECQ subscalesa, mean (SD) Rehearsala Emotional inhibitiona Benign controla Aggression controla CAPS-CAa, mean (SD) Measures C-GAS, mean (SD)

16.7 (1.6) 107.1 (10.7)

16.0 (1.3) 114.0 (8.4)

16.8 (1.2) 111.0 (15.7)

2.02 1.7

n.s. n.s.

9 (45.0) 4 (20.0) 7 (35.0)

13 (65.0) 5 (25.0) 2 (10.00)

10 (50.0) 5 (25.0) 5 (25.0)

– – –

– – –

10 (50.0) 2 (10.0) 47.5 (8.2) 27.3 (12.7) 24.3 (10.6) 3.0 (1.4) 70.6 (13.7)

7 (35.0) 0 61.9 (9.3) 10.5 (8.5) 12.5 (10.2) 1.4 (1.2) 59.0 (8.9)

n.a. n.a. n.a. 3.7 (4.3) 12.0 (16.4) 1.1 (0.9) 55.4 (9.9)

34.97 5.95 15.38 10.41

– – – o 0.001 0.005 o 0.001 o 0.001

23.2 (5.7) 30.8 (4.4) 13.3 (3.0) 27.0 (4.8) 27.2 (13.8) BPD 47.5 (8.2)

18.8 (5.6) 25.9 (5.4) 10.3 (1.6) 21.4 (3.4) 17.1 (11.3) CC 61.9 (9.3)

17.4 (3.5) 26.2 (4.1) 11.1 (2.4) 23.7 (3.1) 14.3 (8.4) HC n.a.

7.28 6.95 7.81 10.58 7.05 T (1,38)  5.45

0.002 0.002 0.001 o 0.001 0.002 P 0.001

School type: Gymnasium, eight years of school after four years of elementary school, terminating with the general qualification for university entrance. Realschule, six years of school after four years of elementary school, terminating with a secondary-school level-I certificate. Hauptschule, nine years of elementary school. Abbreviations: BPD, borderline personality disorder; CC, clinical controls; HC, healthy controls; n.s., not significant; n.a., not available; IQ, intelligence quotient; C-GAS, Children's Global Assessment Scale; BDI, Beck Depression Inventory; BAI, Beck Anxiety Inventory; A-DES, Adolescent Dissociative Experience Scale; BIS, Barratt Impulsiveness Scale; ECQ, Emotional Control Questionnaire; CAPS-CA, Clinician Administered Posttraumatic Stress Disorder Scale, Child and Adolescent version. a

Pairwise post-hoc comparisons of groups: BPD4CC, BPD 4HC, CC ¼ HC.

182

J. Richter et al. / Psychiatry Research: Neuroimaging 221 (2014) 179–186

(FWHM) kernel of 10 mm. Results were corrected for multiple comparisons by false discovery rate (FDR) at Po 0.05 (Genovese et al., 2002). Since FDR is not a very conservative method to correct for multiple comparisons, clusters under a cluster size of 20 voxels are eliminated.

2.5. Statistical analysis of the demographic and psychometric data Comparisons between demographic and psychometric data of the subjects were calculated using chi-square tests for categorical variables and analysis of variance (ANOVA) for continuous variables. An ANCOVA was calculated to control for potentially confounding variables on grey matter volumes within the BPD group. Statistical analysis was done with the statistical software program Stata, version 12 (StataCorp, 2011).

3. Results 3.1. Sample characteristics There were no significant group differences between the group of patients with BPD (n¼ 20), CC (n ¼20), and HC (n ¼ 20) regarding age (F(2,57) ¼ 2.02, P ¼0.141) and IQ (F(2,57) ¼ 1.70, P ¼0.193). Comorbid diagnoses in patients with BPD were mood disorders (n ¼9), anxiety disorders (n ¼9), substance abuse (n ¼9), eating disorders (n ¼7), and conduct disorders (n ¼2). Diagnoses of CC comprised mood disorders (n ¼4), anxiety disorders (n ¼ 4), eating

disorders (n ¼6), somatoform disorders (n ¼3), adjustment disorders (n¼ 6), conduct disorders (n ¼1) and attention deficit/ hyperactivity disorders (ADHD) (n ¼1). Patients could receive more than one psychiatric diagnosis. At the time of the scan, some patients had received medical treatment. Nine of the adolescent girls in the group of patients with BPD were taking psychopharmacological medications, with seven patients taking antidepressants (selective serotonin re-uptake inhibitors (SSRIs) (n ¼6), tricyclic antidepressant (n ¼1), or both SSRIs and a tricyclic antidepressant (n¼ 1)). In the group with CC, two patients were taking SSRIs, one patient was taking a tricyclic antidepressant, and one was taking valproic acid. Girls with BPD showed a significantly impaired psychological functioning as measured by the C-GAS when compared with CC. Pairwise post-hoc comparisons revealed higher scores in all other psychometric scores in adolescents with BPD compared both with CC and HC. Demographic and psychometric results as well as results of post-hoc comparisons are given in Table 1. 3.2. Volumetric data In subcortical structures, the ANCOVA highlighted significant differences in the volume of the right and left hippocampus and significant volumetric differences in the right amygdala (Fig. 1A and B).

Fig. 1. Group differences in subcortical (A and B), frontal (C and D) and parietal (E) structures between subjects with borderline personality disorder (BPD) (n ¼20), clinical controls (CC) (n¼ 20) and healthy controls (HC) (n¼ 20) in mm3.

J. Richter et al. / Psychiatry Research: Neuroimaging 221 (2014) 179–186

Post-hoc comparisons in the right and left hippocampus revealed significant group differences between the patients with BPD and HC, as well as group differences between CC and HC in the right hippocampus. In the right amygdala, the patients with BPD differed significantly from HC. No other group differences were detected. Controlling for comorbid PTSD, medication status, number of diagnoses, and the interaction between medication status and number of diagnoses on limbic grey matter volumes within the BPD group did not reveal any significant results (data available on request). Means, standard deviations, F- and P-values and significant results of post-hoc comparisons are shown in Table 2 and 3; see bar diagrams in Fig. 1. In the frontal lobe, the ANCOVA revealed significant differences in grey matter volume between the three groups in the right middle frontal gyrus and in the orbital part of the inferior frontal gyrus bilaterally (Fig. 1C and D). Post-hoc comparisons between the three groups of each significant neuroanatomical structure were conducted. Post-hoc comparisons revealed that patients with BPD differed from HC in the right middle frontal gyrus and in the orbital part of the left inferior frontal gyrus. Volumetric changes between patients with BPD and CC could be observed in the orbital part of the right inferior frontal gyrus. CC and HC showed altered volumes in the orbital part of the left inferior frontal gyrus.

183

In the parietal lobe, the ANCOVA revealed significant group differences in the right and left superior parietal gyrus (SPG) (Fig. 1E). Post-hoc comparisons revealed a significant difference between the group of subjects with BPD and HC in the right SPG. In the left SPG, significant differences between patients with BPD and HC, as well as between patients with BPD and CC, were found. Means, standard deviations, F- and P-values, and significant differences of post-hoc comparisons are shown in Table 2; see bar diagrams in Fig. 1. 3.3. Cortical thickness analysis No significant differences in cortical thickness among girls with BPD, CC, and HC were detected in the left or in the right hemisphere.

4. Discussion Our results revealed a decreased volume of the hippocampus bilaterally and a volume reduction of the right amygdala in adolescents with BPD when compared with HC. CC differed from HC in the right hippocampus. As comorbid PTSD, medication status, number of diagnoses and the interaction between number

Table 2 Means, standard deviations, F-values and significant P-values (P o.05) of parietal and frontal lobe regions, and limbic structures (in mm3) in subjects with borderline personality disorder (BPD) (n¼ 20), clinical controls (CC) (n¼20) and healthy controls (HC) (n¼ 20). Structure

BPD

Hippocampus Right, mean (SD) 3977.65 Left, mean (SD) 3748.75 Amygdala Right, mean (SD) 1613 Middle frontal gyrus Right, mean (SD) 9472.95 Orbital part of the inferior frontal Right, mean (SD) 855.8 Left, mean (SD) 801.95 Superior parietal gyrus Right, mean (SD) 4823.85 Left, mean (SD) 6137.05

CC

HC

F

(2,57)

P

P—values of post-hoc comparisons BPD vs. HC

BPD vs. CC

CC vs. HC

(70.49) (82.26)

4066.35 (66.47) 3877.5 (96.60)

4339.8 (74.66) 4167.5 (81.87)

6.48 5.24

0.003 0.008

0.002 0.005

n.s. n.s.

0.033 n.s.

(49.58)

1712.45 (33.78)

1777 (38.16)

3.56

0.035

0.024

n.s.

n.s.

(332.40) gyrus (38.52) (30.89)

9937.05 (323.39)

10898.55 (355.48)

4.71

0.013

0.003

n.s.

n.s.

1006.6 (46.46) 1018.35 (46.83)

3.61 7.35

0.034 0.002

n.s. 0.001

0.032 n.s.

n.s. 0.024

(142.20) (128.33)

5397.55 (167.41) 6959.6 (210.76)

5944.15 (244.4) 6984.25 (233.63)

7.59 5.83

0.001 0.005

o 0.001 0.007

n.s. 0.013

n.s. n.s.

1025.4 (57.53) 871.3 (44.85)

Table 3 Comparison of our morphometric/volumetric studies examining adolescents with borderline personality disorder (BPD). Structure Dorsolateral prefrontal cortex Left Right Orbitofrontal cortex Left Medial frontal gyrus Right Orbital part of inferior frontal gyrus Left Right Superior parietal gyrus Left Right Hippocampus Left Right Amygdala Right

VBM (Brunner et al., 2010)

Volume analysis using FreeSurfer

BPDo HC BPDo HC; CC o HC

Segmentation not implemented

BPDo HC

n.s.

n.s.

BPDo HC

n.s. n.s.

BPDo HC BPDo CC

n.s. n.s.

BPDo HC; BPD oCC BPDo HC

n.s. n.s.

BPDo HC BPDo HC; CC o HC

n.s.

BPDo HC

Note: HC, healthy control group; CC, clinical control group; VBM, voxel-based morphometry; n.s., not significant.

184

J. Richter et al. / Psychiatry Research: Neuroimaging 221 (2014) 179–186

of diagnoses and medication status did not have any impact on grey matter volumes within the BPD group, volumetric alterations in limbic structures may be contributed rather to the disorder itself than to other variables. With regard to alterations of the frontal lobe, adolescent girls with BPD demonstrated a decreased grey matter volume in the right frontal middle gyrus and in the orbital part of the left inferior frontal gyrus when compared with HC. Patients with BPD and CC differed from each other in the orbital part of the right inferior frontal gyrus, but both groups did not differ from the HC in this structure. CC differed from the HC in the orbital part of the left inferior frontal gyrus. In parietal areas, adolescent girls with BPD differed both from HC and CC in the left SPG, indicating that the grey matter decrease in this structure might be disorder specific. No differences in limbic structures and parietal areas were found in our earlier VBM study (Brunner et al., 2010). The results of the present study are more in line with our study examining white matter fiber tracks in BPD (Maier-Hein et al., 2014), both pointing towards a more widespread involvement of different brain areas in the disorder. The difference between the results obtained by FreeSurfer in the present study and by SPM5 in the earlier data analysis (Brunner et al., 2010) may reflect the different analytic approaches of the two methods. In SPM5, analyses of the MR images are based on a voxel-wise comparison of the tissue after all MR images are normalized into the same stereotactic space (Ashburner and Friston, 2005, 2000). In contrast to SPM5, FreeSurfer analyses grey matter volumes as structures as a whole without voxel-wise comparisons between the individual MR images (Eggert et al., 2012). Both methods use different processing pipelines and have a varying atlas composition (Dewey et al., 2010). Studies reported that FreeSurfer yields very accurate results with regard to subcortical structures. For this reason, we assume that FreeSurfer is a sensitive method for analyzing limbic structures. Summing up the results regarding limbic structures, the present study is the first study to find decreased grey matter volumes in the hippocampus bilaterally and in the right amygdala in adolescent girls with BPD, indicating that volume alterations in these neuroanatomical structures are already present before adulthood and not simply a correlate of chronicity or long-term medication with psychotropic agents. No study has detected limbic changes in an adolescent sample before (Brunner et al., 2010; Chanen et al., 2008; Goodman et al., 2011). Even our own morphometric study using the same dataset as in the present study only detected volumetric alterations in frontal areas and no difference in any limbic structure, due to a different method of analysis (Brunner et al., 2010). However, the volumetric alterations of the limbic structures are not specific to BPD as the BPD group only differs from HC and not from CC. Neither the study using VBM (Brunner et al., 2010) nor the present study analyzing the same dataset with FreeSurfer showed volumetric specificity in limbic structures. However, in all limbic structures where volumetric differences were revealed, the smallest volume was detected in BPD, followed by CC and HC. As the volumes of CC are always located between the volumes of HC and BPD in the present study, this may indicate that changes in limbic structures display an overall vulnerability for the development of psychiatric disorders, especially emerging in subjects having BPD. Another explanation for the lack of specificity of the results to BPD might be that the source of variance is not yet detected. The subjects of our BPD group exhibit a high amount of comorbid disorders that may have an impact on the volume of brain structures. For example, as already mentioned, the volume of the amygdala in BPD patients is influenced by the presence or absence of comorbid PTSD (de-Almeida et al., 2012).

However, one has to take into account that an important characteristic of BPD is having a large number of comorbid disorders. Even though not all patients with BPD show a comorbid diagnosis in the present study, additional symptoms to BPD may appear in a dimensional way not reaching the cut-off criteria of other diagnoses. The presence of comorbid disorders in almost all subjects with BPD is affirmed by a study investigating comorbid axis I and axis II disorders in patients with BPD in comparison to a mixed psychiatric sample (without BPD diagnosis) (Kaess et al., 2013). It revealed that the BPD sample showed significantly higher rates of both axis I and axis II disorders. Furthermore, none of the patients with BPD displayed an isolated BPD diagnosis, all having at least one more comorbid diagnosis (Kaess et al., 2013). If these findings are borne in mind, a study representing exclusively patients with symptoms of BPD without any comorbid disorder would not reflect the clinical picture of this personality disorder and would reduce the comparability and the generalizability of the results. Regarding the results of the parietal lobule, the specificity to BPD of the left SPG is of great importance. To the authors’ knowledge, only three studies dealt with volume alterations in parietal lobe structures in subjects with BPD, with none of them dealing with an adolescent sample (Irle et al., 2005, 2007; Völlm et al., 2009). Further investigations are needed if a decreased volume in this structure constitutes vulnerability for developing BPD or if volume changes in the left SPG develop due to environmental factors or comorbid disorders. The aim of the examination of cortical thickness in adolescents with BPD was to gain a deeper insight into the underlying mechanisms responsible for volumetric changes of cortical structures. Comparing cortical thickness between girls with BPD, CC and HC revealed no significant differences among the groups, indicating that adolescents with BPD do not have altered cortical structures relative to other adolescents in their age range. As grey matter volume is determined by both cortical thickness and cortical surface area (Hutton et al., 2008; Panizzon et al., 2009; Winkler et al., 2010), the results of the present study point towards the possibility that volumetric changes between the groups are generated by changes in cortical surface area. This issue needs further investigation. 4.1. Strengths and limitations One of the major strengths of this study is the fact that none of the patients had been ill for a long time, so the influence of medication, chronicity-associated processes, or long-lasting comorbid disorders is reduced and any structural brain changes may be rather due to the disorder itself than due to other variables. Moreover, patients with BPD are not only compared to a group of HC, but also to a group of CC that consisted of adolescent patients with other psychiatric disorders. To the authors' knowledge, the present study is the only one that includes such a group of CC. There are some limitations of the study. With regard to the composition of the sample, subjects belonging to the BPD group often exhibit other psychiatric disorders besides BPD. The results obtained cannot be interpreted as being only due to the BPD diagnosis, since they could also have been caused by another comorbid psychiatric disorder. This issue is difficult to resolve because most patients with BPD have at least one other psychiatric diagnosis, and having a comorbid disorder is characteristic for the clinical picture of BPD (Kaess et al., 2013). The design of the present study is cross-sectional. No statements can be made on the causality between the observed neuroanatomical alterations and BPD or other psychiatric disorders.

J. Richter et al. / Psychiatry Research: Neuroimaging 221 (2014) 179–186

4.2. Conclusion The main issue of the present article was the investigation of limbic structures, namely the hippocampus and the amygdala, in adolescents with BPD. The dataset used has been analyzed by SPM5 at an earlier date, but FreeSurfer is claimed to be more accurate than SPM for the analysis of subcortical structures (Tae et al., 2008; Dewey et al., 2010). As expected, FreeSurfer detected changes in the limbic system that were not found by SPM5. The present study is the first one to show amygdalar and hippocampal alterations already present in an adolescent sample with BPD. Investigating adolescents is beneficial because neuroanatomical changes in early stages of the disorder are most likely due to the disorder itself and the effects of long-term treatment with psychotropic agents, chronicity, or persistent comorbid disorders are accordingly reduced. A volume decrease in the left SPG is disorder specific, as female adolescents with BPD differed from both CC and HC in this neuroanatomical structure. Further investigations are needed if a decreased volume in this structure constitutes vulnerability for developing BPD. Our study points to no specific changes in cortical thickness in adolescent girls with BPD when compared with CC and HC.

Acknowledgements We thank Dr. Jasmin Kramer, Dr. Nina Feigl, and Dr. Kira Lutz for supporting patient recruitment. We report no biomedical financial interests or potential conflicts of interest. References American Psychiatric Association, 2000. Diagnostic and Statistical Manual of Mental Disorders (DSM) IV, third ed. APA, Washington, DC Ashburner, J., Friston, K.J., 2000. Voxel-based morphometry—the methods. NeuroImage 11, 805–821. Ashburner, J., Friston, K.J., 2005. Unified segmentation. NeuroImage 26, 839–851. Beck, A.T., 1995. Beck's Depressions-Inventar (BDI). Dt. Bearbeitung von M. Hautzinger, M. Bailer, H. Worall F. Keller (Eds.), Hogrefe, Göttingen. Brunner, R., Henze, R., Parzer, P., Kramer, J., Feigl, N., Lutz, K., Essig, M., Resch, F., Stieltjes, B., 2010. Reduced prefrontal and orbitofrontal gray matter in female adolescents with borderline personality disorder: is it disorder specific? NeuroImage 49, 114–120. Buchanan, R.W., Francis, A., Arango, C., Miller, K., Lefkowitz, D.M., McMahon, R.P., Barta, P.E., Pearlson, G.D., 2004. Morphometric assessment of the heteromodal association cortex in schizophrenia. American Journal of Psychiatry 161, 322–331. Carrasco, J.L., Tajima-Pozo, K., Díaz-Marsá, M., Casado, A., López-Ibor, J.J., Arrazola, J., Yus, M., 2012. Microstructural white matter damage at orbitofrontal areas in borderline personality disorder. Journal of Affective Disorders 139, 149–153. Caviness , V.S., Filipek, P.A., Kennedy, D.N., 1989. Magnetic resonance technology in human brain science: blueprint for a program based upon morphometry. Brain Development 11, 1–13. Chanen, A.M., Velakoulis, D., Carison, K., Gaunson, K., Wood, S.J., Yuen, H.P., Yücel, M., Jackson, H.J., McGorry, P.D., Pantelis, C., 2008. Orbitofrontal, amygdala and hippocampal volumes in teenagers with first-presentation borderline personality disorder. Psychiatry Research: Neuroimaging 163, 116–125. Clarkson, M.J., Cardoso, M.J., Ridgway, G.R., Modat, M., Leung, K.K., Rohrer, J.D., Fox, N.C., Ourselin, S., 2011. A comparison of voxel and surface based cortical thickness estimation methods. NeuroImage 57, 856–865. Dahl, G., 1986. Reduzierter Wechsler-Intelligenztest. Hogrefe, Göttingen, Bern Dale, A.M., Fischl, B., Sereno, M.I., 1999. Cortical surface-based analysis: I. Segmentation and surface reconstruction. NeuroImage 9, 179–194. Delmo, C., Weiffenbach, O., Gabriel, M., Poustka, F., 2000. Kiddie-Sads-Present ad Lifetime Version (K-SADS-PL), third ed. Klinik für Psychiatrie und Psychotherapie des Kindes- und Jugendalters, Frankfurt am Main. Dewey, J., Hana, G., Russell, T., Price, J., McCaffrey, D., Harezlak, J., Sem, E., Anyanwu, J.C., Guttmann, C.R., Navia, B., Cohen, R., Tate, D.F., 2010. Reliability and validity of MRI-based automated volumetry software relative to auto-assisted manual measurement of subcortical structures in HIV-infected patients from a multisite study. NeuroImage 51, 1334–1344. Driessen, M., Herrmann, J., Stahl, K., Zwaan, M., Meier, S., Hill, A., Osterheider, M., Petersen, D., 2000. Magnetic resonance imaging volumes of the hippocampus and the amygdala in women with borderline personality disorder and early traumatization. Archives of General Psychiatry 57, 1115–1122.

185

Eggert, L.D., Sommer, J., Jansen, A., Kircher, T., Konrad, C., 2012. Accuracy and reliability of automated gray matter segmentation pathways on real and simulated structural magnetic resonance images of the human brain. PLoS One 7, e45081. First, M.B., Gibbon, M., Spitzer, R.L., Williams, J.B.W., Benjamin, L., 1997. Structured Clinical Interview for DSM-IV Axis II Personality Disorders (SCID-II). American Psychiatric Press, Washington, DC Fischl, B., Dale, A.M., 2000. Measuring the thickness of the human cerebral cortex from magnetic resonance images. Proceedings of the National Academy of Sciences of the United States of America 97, 11050–11055. Fischl, B., Sereno, M.I., Tootell, R.B.H., Dale, A.M., 1999. High-resolution intersubject averaging and a coordinate system for the cortical surface. Human Brain Mapping 8, 272–284. Fischl, B., Salat, D.H., Busa, E., Albert, M., Dieterich, M., Haselgrove, C., van der Kouwe, A., Killiany, R., Kennedy, D., Klaveness, S., Montillo, A., Makris, N., Rosen, B., Dale, A.M., 2002. Whole brain segmentation: automated labeling of neuroanatomical structures in the human brain. Neuron 33, 341–355. Fischl, B., van der Kouwe, A., Destrieux, C., Halgren, E., Ségonne, F., Salat, D.H., Busa, E., Seidman, L.J., Goldstein, J., Kennedy, D., Caviness, V., Makris, N., Rosen, B., Dale, A.M., 2004. Automatically parcellating the human cerebral cortex. Cerebral Cortex 14, 11–22. Fydrich, T., Renneberg, B., Schmitz, B., Wittchen, H.-U., 1997. SKID II. Strukturiertes Klinisches Interview für DSM-IV, Achse II. Persönlichkeitsstörungen. Interviewheft. Eine deutschsprachige, erw. Bearb. d. amerikanischen Originialversion des SKID II. Hogrefe, Göttingen. Genovese, C.R., Lazar, N.A., Nichols, T., 2002. Thresholding of statistical maps in functional neuroimaging using the false discovery rate. NeuroImage 15, 870–878. Goodman, M., Hazlett, E.A., Avedon, J.B., Siever, D.R., Chu, K.-W., New, A.S., 2011. Anterior cingulate volume reduction in adolescents with borderline personality disorder and co-morbid major depression. Journal of Psychiatric Research 45, 803–807. Grant, J., Correia, S., Brennan-Krohn, T., Malloy, P., Laidlaw., D., Schulz, S., 2007. Frontal white matter integrity in borderline personality disorder with selfinjurious behavior. Journal of Neuropsychiatry and Clinical Neurosciences 19, 383–390. Hazlett, E.A., New, A.S., Newmark, R., Haznedar, M.M., Lo, J.N., Speiser, L.J., Chen, A. D., Mitropoulou, V., Minzenberg, M., Siever, L.J., Buchsbaum, M.S., 2005. Reduced anterior and posterior cingulate gray matter in borderline personality disorder. Biological Psychiatry 58, 614–623. Hutton, C., De Vita, E., Ashburner, J., Deichmann, R., Turner, R., 2008. Voxel-based cortical thickness measurements in MRI. NeuroImage 40, 1701–1710. Irle, E., Lange, C., Sachsse, U., 2005. Reduced size and abnormal asymmetry of parietal cortex in women with borderline personality disorder. Biological Psychiatry 57, 173–182. Irle, E., Lange, C., Weniger, G., Sachsse, U., 2007. Size abnormalities of the superior parietal cortices are related to dissociation in borderline personality disorder. Psychiatry Research: Neuroimaging 156, 139–149. Kaess, M., von Ceumern-Lindenstjerna, I.-A., Parzer, P., Chanen, A., Mundt, C., Resch, F., Brunner, R., 2013. Axis I and II comorbidity and psychosocial functioning in female adolescents with borderline personality disorder. Psychopathology 46, 55–62. Kaufman, J., Birmaher, B., Brent, D., Rao, U., Flynn, C., Moreci, P., Williamson, D., Ryan, N., 1997. Schedule for affective disorders and schizophrenia for schoolage children-present and lifetime version (K-SADS-PL): initial reliability and validity data. Journal of the American Academy of Child and Adolescent Psychiatry 36, 980–988. Leichsenring, F., Leibing, E., Kruse, J., New, A., Leweke, F., 2011. Borderline personality disorder. Lancet 377, 74–81. Lis, E., Greenfield, B., Henry, M., Guilé, J.M., Dougherty, G., 2007. Neuroimaging and genetics of borderline personality disorer: a review. Journal of Psychiatry and Neuroscience 32, 162–173. Lyoo, I.K., Han, M.H., Cho, D.Y., 1998. A brain MRI study in subjects with borderline personality disorder. Journal of Affective Disorders 50, 235–243. Maier-Hein, K.H., Brunner, R., Lutz, K., Henze, R., Parzer, P., Feigl, N., Kramer, J., Meinzer, H.-P., Resch, F., Stieltjes, B., 2014. Disorder-specific white matter alterations in adolescent borderline personality disorder. Biological Psychiatry 75 (1), 81–88. Minzenberg, M.J., Fan, J., New, A.S., Tang, C.Y., Siever, L.J., 2008. Frontolimbic structural changes in borderline personality disorder. Journal of Psychiatric Research 42, 727–733. Nader, K.O., Kriegler, J.A., Blake, D.D., Pynoos, R.S., Newman, E., 1996. Clinician Administered PTSD Scale, Child and Adolescent Version (CAPS-CA). National Center for PTSD, White River Junction, VT New, A.S., Carpenter, D.M., Perez-Rodriguez, M.M., Ripoll, L.H., Avedon, J., Patil, U., Hazlett, E.A., Goodman, M., 2013. Developmental differences in diffusion tensor imaging parameters in borderline personality disorder. Journal of Psychiatric Research 47, 1101–1109. Niedtfeld, I., Schulze, L., Krause-Utz, A., Demirakca, T., Bohus, M., Schmahl, C., 2013. Voxel-based morphometry in women with borderline personality disorder with and without comorbid posttraumatic stress disorder. PLoS One 8, e65824. Nunes, P.M., Wenzel, A., Borges, K.T., Porto, C.R., Caminha, R.M., de Oliveira, I.R., 2009. Volumes of the hippocampus and amygdala in patients with borderline personality disorder: a meta-analysis. Journal of Personality Disorders 23, 333–345. Oldfield, R.C., 1971. The assessment and analysis of handedness: the Edinburgh Inventory. Neuropsychologia 9, 97–113.

186

J. Richter et al. / Psychiatry Research: Neuroimaging 221 (2014) 179–186

Panizzon, M.S., Fennema-Notestine, C., Eyler, L.T., Jernigan, T.L., Prom-Wormley, E., Neale, M., Jacobson, K., Lyons, M.J., Grant, M.D., Franz, C.E., Xian, H., Tsuang, M., Fischl, B., Seidman, L., Dale, A., Kremen, W.S., 2009. Distinct genetic influences on cortical surface area and cortical thickness. Cerebral Cortex 19, 2728–2735. Paris, J., 2005. Borderline personality disorder. Canadian Medical Association Journal 172, 1579–1583. Rodrigues, E., Wenzel, A., Ribeiro, M.P., Quarantini, L.C., Miranda-Scippa, A., de Sena, E.P., de Oliveira, I.R., 2011. Hippocampal volume in borderline personality disorder with and without comorbid posttraumatic stress disorder: a metaanalysis. European Psychiatry 26, 452–456. Ross, C.A., Pearlson, G.D., 1996. Schizophrenia, the heteromodal association neocortex and development: potential for a neurogenetic approach. Trends in Neuroscience 19, 171–176. Ruocco, A.C., Amirthavasagam, S., Zakzanis, K.K., 2012. Amygdala and hippocampal volume reductions as candidate endophenotypes for borderline personality disorder: a meta-analysis of magnetic resonance imaging studies. Psychiatry Research: Neuroimaging 201, 245–252. Rüsch, N., Tebartz van Elst, L., Ludaescher, P., Wilke, M., Huppertz, H.-J., Thiel, T., Schmahl, C., Bohus, M., Lieb, K., Heßlinger, B., Hennig, J., Ebert, D., 2003. A voxel-based morphometric MRI study in female patients with borderline personality disorder. NeuroImage 20, 385–392. Rüsch, N., Bracht, T., Kreher, B.W., Schnell, S., Glauche, V., Il'yasov, K.A., Ebert, D., Lieb, K., Hennig, J., Saur, D., van Elst, L.T., 2010. Reduced interhemispheric structural connectivity between anterior cingulate cortices in borderline personality disorder. Psychiatry Research: Neuroimaging 181, 151–154. Schmahl, C., Bremner, D., 2006. Neuroimaging in borderline personality disorder. Journal of Psychiatric Research 40, 419–427. Schmahl, C.G., Vermetten, E., Elzinga, B.M., Douglas Bremner, J., 2003. Magnetic resonance imaging of hippocampal and amygdala volume in women with childhood abuse and borderline personality disorder. Psychiatry Research: Neuroimaging 122, 193–198. Selby, E.A., Anestis, M.D., Bender, T.W., Joiner , T.E., 2009. An exploration of the emotional cascade model in borderline personality disorder. Journal of Abnormal Psychology 118, 375–387. Shaffer, D., Gould, M.S., Brasic, J., Ambrosini, P., Fisher, P., Bird, H., Aluwahlia, S., 1983. A children's global assessment scale (CGAS). Archives of General Psychiatry 40, 1228–1231. StataCorp, 2011. Stata Statistical Software. Release 12. StataCorp LP, College Station, TX Tae, W., Kim, S., Lee, K., Nam, E.-C., Kim, K., 2008. Validation of hippocampal volumes measured using a manual method and two automated methods

(FreeSurfer and IBASPM) in chronic major depressive disorder. Neuroradiology 50, 569–581. Takahashi, T., Chanen, A.M., Wood, S.J., Walterfang, M., Harding, I.H., Yücel, M., Nakamura, K., McGorry, P.D., Suzuki, M., Velakoulis, D., Pantelis, C., 2009. Midline brain structures in teenagers with first-presentation borderline personality disorder. Progress in Neuro-Psychopharmacology and Biological Psychiatry 33, 842–846. Tebartz van Elst, L., Hesslinger, B., Thiel, T., Geiger, E., Haegele, K., Lemieux, L., Lieb, K., Bohus, M., Hennig, J., Ebert, D., 2003. Frontolimbic brain abnormalities in patients with borderline personality disorder: a volumetric magnetic resonance imaging study. Biological Psychiatry 54, 163–171. Völlm, B.A., Zhao, L., Richardson, P., Clark, L., Deakin, J.F.W., Williams, S., Dolan, M.C., 2009. A voxel-based morphometric MRI study in men with borderline personality disorder: preliminary findings. Criminal Behavior and Mental Health 19, 64–72. Walhovd, K.B., Fjell, A.M., Reinvang, I., Lundervold, A., Dale, A.M., Eilertsen, D.E., Quinn, B.T., Salat, D., Makris, N., Fischl, B., 2005. Effects of age on volumes of cortex, white matter and subcortical structures. Neurobiological of Aging 26, 1261–1270. Wechsler, D., 1999. Wechsler Abbreviated Scale of Intelligence. Psychological Cooperation, San Antonio, TX Weniger, G., Lange, C., Sachsse, U., Irle, E., 2009. Reduced amygdala and hippocampus size in trauma-exposed women with borderline personality disorder and without posttraumatic stress disorder. Journal of Psychiatry and Neuroscience 34, 383–388. Whittle, S., Chanen, A.M., Fornito, A., McGorry, P.D., Pantelis, C., Yücel, M., 2009. Anterior cingulate volume in adolescents with first-presentation borderline personality disorder. Psychiatry Research: Neuroimaging 172, 155–160. Winkler, A.M., Kochunov, P., Blangero, J., Almasy, L., Zilles, K., Fox, P.T., Duggirala, R., Glahn, D.C., 2010. Cortical thickness or grey matter volume? The importance of selecting the phenotype for imaging genetics studies. NeuroImage 53, 1135–1146. Zetzsche, T., Frodl, T., Preuss, U.W., Schmitt, G., Seifert, D., Leinsinger, G., Born, C., Reiser, M., Möller, H.-J., Meisenzahl, E.M., 2006. Amygdala volume and depressive symptoms in patients with borderline personality disorder. Biological Psychiatry 60, 302–310. de-Almeida, C.P., Wenzel, A., de-Carvalho, C.S., Powell, V.B., Araújo-Neto, C., Quarantini, L.C., de-Oliveira, I.R., 2012. Amygdalar volume in borderline personality disorder with and without comorbid post-traumatic stress disorder: a meta-analysis. CNS Spectrums 17, 70–75.