Journal of Affective Disorders 120 (2010) 54–61

Contents lists available at ScienceDirect

Journal of Affective Disorders

j ourna l homepage: www.e lsev ie r.com/ locate / j ad

Research report

Prefrontal-temporal gray matter deficits in bipolar disorder patients withpersecutory delusions

Heike Tost a,b,⁎, Matthias Ruf b, Christine Schmäl a, Thomas G. Schulze a, Carolin Knorr a,Christian Vollmert c, Katja Bößhenz a, Gabriele Ende b, Andreas Meyer-Lindenberg d,Fritz A. Henn d,e, Marcella Rietschel a

a Division of Genetic Epidemiology in Psychiatry, Central Institute of Mental Health, J5, 68159 Mannheim, Germanyb Department Neuroimaging, Central Institute of Mental Health, J5, 68159 Mannheim, Germanyc Department of Addictive Behavior and Addiction Medicine, Central Institute of Mental Health, J5, 68159 Mannheim, Germanyd Department of Psychiatry, Central Institute of Mental Health, J5, 68159 Mannheim, Germanye Brookhaven National Laboratory, Upton, NY 11973-5000, United States

a r t i c l e i n f o

⁎ Corresponding author. Department NeuroimaginMental Health, Square J5, 68159 Mannheim, Germany

E-mail address: heike.tost@zi-mannheim.de (H. To

0165-0327/$ – see front matter © 2009 Elsevier B.V.doi:10.1016/j.jad.2009.04.009

a b s t r a c t

Article history:Received 24 September 2008Received in revised form 17 March 2009Accepted 7 April 2009Available online 5 May 2009

Objectives: Although brain structural deficits have been repeatedly associated with bipolardisorder (BD), inconsistency in morphometric results has been a feature of neuroimagingstudies. We hypothesize that this discrepancy is related to the heterogeneity of BD, andexamine the question of whether or not more homogeneous clinical subgroups display a morecoherent pattern of morphometric abnormalities.Methods: In a case-control design, we examined differences in gray matter (GM), white matter(WM) and cerebrospinal fluid (CSF) concentration in 42 BD patients and 42 healthy matchedcontrols using optimized voxel-basedmorphometry (VBM). Subgroup analyses of patients witha lifetime history of psychotic symptoms (BDP, n=30) and patients with mood-incongruentpsychotic symptoms in the form of persecutory delusions (BDPD, n=15) were performed toaccord with previous genetic findings.Results: Analysis of the total BD samplewas largely inconclusive, but the BDPD patient subgroupdisplayed awidespread pattern of significant decreases in GM concentration in the dorsolateralprefrontal (DLPFC), temporal and cingulate cortices, and a significant CSF increase in theadjacent outer ventricular sulci. Comparison of BDPD patients versus BD and BDP patientswithout persecutory delusions revealed a significant GM decrease in the left DLPFC for theformer group.Conclusions: BDPD show pronounced structural abnormalities of the prefrontal and temporallobes which are similar to the deficits previously reported for schizophrenia (SCZ). Our findingssuggest that stratification based solely on psychotic symptoms is insufficient for thedifferentiation of BD into biologically meaningful subgroups, but also question thepathophysiological validity of the dichotomy in classification between schizophrenia and BD.

© 2009 Elsevier B.V. All rights reserved.

Keywords:Bipolar disorderPersecutory delusionsVoxel-based morphometryGray matterCerebrospinal fluidSchizophrenia

1. Introduction

Bipolar disorder (BD) is a genetically and clinically het-erogeneous condition affecting approximately 1.5% of the

g, Central Institute o.st).

All rights reserved.

f

population (Regeer et al., 2004). Although substantial effortshave been made to elucidate its origin, the precise neurobio-logical underpinnings of the disorder are still unknown.Previous functional imaging work has implicated anomaliesin the “visceromotor” network of the brain (Ongur and Price,2000), a set of highly interconnected cortical and subcorticalareas involved in the processing of emotionally salient stimuli(i.e., prefrontal cortex, subgenual anterior cingulate cortex,

55H. Tost et al. / Journal of Affective Disorders 120 (2010) 54–61

amygdala, hippocampus, striatum; see Strakowski et al., 2005for a comprehensive review). Associated brain structural al-terations have been described repeatedly in BD. Previousstructural magnetic resonance imaging (MRI) studies haveidentified subtle gray matter (GM) volume differences inseveral submodules of the visceromotor network (particularlythe amygdala), although the direction of these differenceswith respect to healthy controls is contentious (Strakowskiet al., 2005). Recent evidence suggests that gray mattervolume increases in the anterior cingulate gyrus (ACG) arepresent early in the course of illness,while atrophic changes ofthe prefrontal cortex (PFC) develop as a function of illnesschronicity (Adler et al., 2007). However, due to the substantialamount of negative and conflicting results current scientificknowledge of brain morphometric deficits in BD remainsinconclusive (McDonald et al., 2004). Previous brain morpho-metric research in patients with schizophrenia (SCZ) hasproduced amore coherent picture (seeWright et al., 2000 for acomprehensive review). A recent meta-analysis of VBMstudies identified GM reductions in the bilateral superiortemporal gyrus, left medial temporal lobe, and left medial andinferior frontal gyrus asmost consistent findings (Honea et al.,2005). Both, longitudinal structural imaging studies (vanHaren et al., 2007) and post-mortem studies (Harrison, 1999)in SCZ suggest the involvement of a chronic disease processthat preferentially affects the prefrontal and temporal lobes.

Experience from other complex neuropsychiatric disor-ders would suggest that the examination of more homo-geneous patient subgroups is the most promising approach inidentifying brain structural abnormalities associated withparticular aspects of the disease. In accordance with thisapproach, previous morphometric studies have focused onthe examination of bipolar disorder patients with a life-timehistory of psychotic symptoms (BDP), a subgroup that showssubstantial overlap with the clinical and neuropsychologicalmanifestations of SCZ (Glahn et al., 2007). None of thesefindings have been consistently replicated (McDonald et al.,2006), however, suggesting that refining samples merely interms of psychotic symptomsmay still be too crude ameasurefor defining a biologically meaningful subtype of BD.

In a subgroup of BDP patients (30%–40%), the content ofpsychotic symptoms is not consistent with traditionaldepressive or manic themes such as guilt, catastrophe,grandiosity or overstated power (Black and Nasrallah, 1989).Recent molecular genetic evidence suggests that it is thephenotypic manifestation of mood-incongruent psychosis inBD that is indicative of a more severe form of the illness(Harrow et al., 2000), and a shared pathophysiology with SCZ(Goes et al., 2007) (Green et al., 2005). Support comes fromone of our own recent molecular genetics studies of the G72/G30 locus (Schulze et al., 2005), one of the best replicatedgenetic loci in schizophrenia (Abou Jamra et al., 2006). Theobserved genotype-phenotype association was found to berestricted to a specific subsample of BDP patients with mood-incongruent psychosis — the individuals with a positive life-time history of persecutory delusions (BDPD). Previousfunctional imaging data suggest that deficits in the medialprefrontal cortex (MedPFC) (Blackwood et al., 2004; Sabriet al., 1997) and dorsolateral prefrontal cortex (DLPFC)(Corlett et al., 2007) are at the core of the development ofpersecutory delusions.

In this study, we addressed the question of whether BDpatients with persecutory delusions also distinguish them-selves from other BD patients with respect to structuralneuroimaging parameters. Employing voxel-based morpho-metry we hypothesized that:

1) Compared to healthy controls, BDPD patients display a morepronounced and coherent pattern of brain structural abnorm-alities than a broader and less refined patient subgroup thatsatisfies the DSM IV criteria for either BD or BDP.

2) The observedmorphometric deficits in BDPD are similar tothe frontal-temporal atrophy pattern previously describedin SCZ, and

3) Compared to BD patients without persecutory delusions,BDPD patients show a significant reduction in gray matterdensity in the prefrontal cortex.

2. Methods

2.1. Subjects

Forty-two patients with bipolar I disorder (19 males, 23females, mean age=42.4, SD=13.1 years) were recruitedfrom consecutive inpatient admissions to the Central Instituteof Mental Health (CIMH) in Mannheim, Germany, fromJanuary 2004 to December 2005. Forty-two healthy commu-nity volunteers with no history of major psychiatric disorderwere group-matched with patients for age and sex (19 males,23 females, mean age=42.2, SD=13.5 years). None of theparticipants had a history of any significant general medical orneurological disorder or head trauma. The study wasapproved by the local University Ethics Committee.

2.2. Diagnostic evaluation

DSM-IV diagnoses (in bipolar I disorder patients) or theirabsence (in healthy controls) were confirmed using theStructured Clinical Interview for DSM-IV disorders (First et al.,2002). At the time of the study, all healthy subjects were freeof any Axis I psychiatric condition with the exception ofnicotine abuse or dependence. In bipolar disorder patients,homogenous clinical subgroups were defined using theOPCRIT system (McGuffin et al., 1991). A broader psychoticsubsample with a positive lifetime history of hallucinations ordelusions (BDP, n=30) and a more circumscribed patientsubsample with persecutory delusions (BDPD, n=15) weredelineated. Eight of the 42 BD subjects had a positive life-timehistory of hallucinations (n=8 subjects in the BDP group, andn=5 subjects in the BDPD group). The Young Mania RatingScale (YMRS) (Young et al., 2000) was administered todetermine the severity of manic and depressive symptoms atadmission. Premorbid intelligence was assessed with amultiple choice verbal intelligence test (Mehrfachwahl–Wortschatz–Intelligenztest MWTB) (Lehrl, 2005). All diag-nostic and psychometric evaluations were performed byexperienced clinical raters.

2.3. Image acquisition

Magnetic resonance imaging scans were performed at theCIMHona1.5TSiemensMagnetomVisionMRscanner (Siemens

56 H. Tost et al. / Journal of Affective Disorders 120 (2010) 54–61

Medical Solutions, Erlangen, Germany) following a standardizedresearch protocol. High-resolution structural images wereobtained with a three-dimensional, T1-weighted MPRAGEsequence with whole-brain coverage (162 slices, 1 mm slicethickness, 256mmfield of view,1mm3 resolution). In all bipolardisorder patients, MRI data acquisition was performed in theremission phase once a euthymic mood state had been reached.

2.4. Image processing

Automated image processing was performed with thesoftware package SPM2 (http://www.fil.ion.ucl.ac.uk/spm/)running in Matlab 6.5 (The MathWorks, Nattick, MA, USA)using an optimized VBM protocol (Good et al., 2001). Imageswere segmented in native space into gray matter (GM), whitematter (WM) and cerebrospinal fluid (CSF) partitions andspatially normalized to the MNI152 (Montreal NeurologicalInstitute) template using a 12-parameter affine transformation.All native images were recursively segmented into tissuecompartments and normalized to the customized template.Resulting tissue partitionswere smoothedwith a 12mmFWHMisotropic Gaussian kernel to reduce the impact of inter-individual variance inbrainmorphologyand satisfy theGaussiandistribution assumptions of subsequent parametric analyses.

2.5. Statistical analysis

Between-group comparisons of demographic and clinicalvariables were performed as follows. Categorical variableswere compared by means of chi-square tests (age, gender,medication status, family history of schizophrenia). Interval-scaled variables were compared using Student's t-test forindependent samples, with the exception of illness duration,illness severity and number of manic and depressive episodes(Multiple Analysis of Covariance MANCOVA with age and

Table 1Demographical and clinical characteristics. a

Bipolar disorder (BD)

BD b (n=42) BDP c (n

Age (years) 42.4±13.1 42.7±1Gender (% women) 54.8 56.7Education (years) 10.8±2.1 10.7±2Age onset (years) 27.5±11.8 28.1±1Illness duration (years) 15.9±9.0 15.4±9Illness severity e 29.2±7.9 27.4±7.Number of admissions f 0.30±0.50 0.34±0# Depressive episodes 4.8±4.5 5.1±5.1# Manic episodes 5.2±4.9 6.0±5.5Medication status (% on)

Total 88.1 90.0Phase prophylaxis 69.0 66.7Antidepressants 28.6 23.3Antipsychotics 38.1 50.0Anxiolytics 14.3 10.0

YMRS (total score) 11.4±15.4 13.1±16MWTB IQ 114.9±13.8 113.7±1Family history SCZ (%) 8.3 20.0

a Values indicate mean±standard deviation unless otherwise specified.b Patients satisfying DSM-IV criteria for bipolar disorder.c Subgroup of BD patients with psychotic features.d Subgroup of psychotic BD patients with a lifetime history of persecutory delusie Lowest score on the Global Assessment of Functioning (GAF) scale during episof Individual number of admissions, weighted by illness duration.

gender as covariates of no interest). Regional differences inGM,WMand CSFwere examinedwithin the framework of thegeneral linear model with a random-effects group statistics atthe second level in SPM2 (Friston et al., 1995). T-tests forindependent samples were used to compare age- and gender-matched groups (total BD sample versus healthy controls).Analysis of Covariance (ANCOVA) with diagnosis as para-meter of interest and age as a covariate of non-interest wasused to compare bipolar disorder subgroups as well as bipolardisorder subgroups and healthy controls. Resulting statisticalmaps were thresholded for t≥5.1, corresponding to a whole-brain corrected family-wise error (FWE) of pb0.05 for allperformed patient-control comparisons at the voxel level.Between-group comparisons of patient subgroups werehypothesis-driven and restricted to an a priori defined regionof interest (ROI) covering the lateral and medial prefrontallobe (Brodmann areas BA 06-12, BA 44-47) and the anteriorcingulate gyrus (BA 31, 24) created with the WFU PickAtlasutility (Maldjian et al., 2003). For the ROI analysis thesignificance level was set at a p value of pb0.0001(t≥4.09), with a minimum cluster size (kE) of 50 voxels.

3. Results

3.1. Demographic and clinical variables

Table 1 summarizes the demographic and clinical charac-teristics of the subject groups examined. None of the bipolardisorder patient groupsdiffered significantly (pb0.05) from thecontrol group in terms of age, gender, or education. Thedifferent BD subsamples did not differ significantly from eachother in terms of demographic characteristics, number ofpatients on psychotropic medication, prescribed substanceclass, or family history of schizophrenia. Similarly, the compar-ison of continuous clinical variables (i.e., age onset, illness

Controls (n=42)

=30) BDPD d (n=15)

3.0 46.6±11.9 42.2±13.653.3 54.8

.1 10.3±1.9 10.9±2.02.1 27.9±7.5 –

.2 19.8±9.1 –

9 27.0±6.4 –

.55 0.36±0.71 –

7.3±6.7 –

8.5±6.2 –

86.6 –

66.7 –

6.7 –

66.7 –

6.7 –

.0 15.5±17.8 –

2.1 115.3±13.0 –

26.7 –

ons.des.

Table 2Regional gray matter density decreases in bipolar disorder.

Patient subgroup Hemisphere BA a Talairach coordinates a Cluster size t-value a p-value a, b

x y z

Temporal lobeSuperior BD c – – – – – – – –

BDP d right 22 60 2 −2 23 5.3 0.030BDPD e right 22 56 12 −6 241 5.9 0.009

Middle BD left 21 −59 0 −20 72 5.2 0.017BDP left 21 −57 0 −24 300 6.1 0.002

right 21 64 −12 −4 34 5.4 0.024BDPD left 21 −56 −1 −19 570 7.2 0.0001

right 21 56 −8 −13 30 5.5 0.028Inferior BD – – – – – – – –

BDP – – – – – – – –

BDPD right 20 53 −9 −32 264 6.4 0.002Prefrontal lobe

Dorsolateral BD – – – – – – – –

BDP – – – – – – – –

BDPD right 9 58 10 27 60 5.7 0.016right 9 48 21 37 240 6.1 0.004left 46 −39 50 20 30 5.6 0.020

Ventrolateral BD – – – – – – – –

BDP right 47 42 26 −2 63 5.4 0.023left 47 −49 27 −10 10 5.2 0.036

BDPD right 45 54 23 2 95 5.6 0.017left 47 −51 20 2 2660 7.1 0.0001

Cingulate lobe BD – – – – – – – –

BDP right & left 31 −17 −22 42 77 5.5 0.016BDPD right & left 31 12 −20 42 89 5.9 0.008

a Local maximum, BA=Brodmann area.b Whole-brain corrected for multiple comparisons (FWE) at the voxel level.c Patients satisfying DSM-IV criteria for bipolar disorder (relative to controls).d Subgroup of BD patients with psychotic features (relative to controls).e Subgroup of psychotic BD patients with a lifetime history of persecutory delusions (relative to controls).

57H. Tost et al. / Journal of Affective Disorders 120 (2010) 54–61

duration, number of admissions, illness severity, premorbid IQ)yielded no significant between-group differences.

3.2. Global brain volumetric measures

Statistical comparison of the total BD patient group andhealthy controls yielded a significant difference for total GM(F=12.84, pb0.001) and CSF volume (F=9.52, pb0.003). Post-hoc comparisons confirmed a significant GM decrease (t=3.01,pb0.003) and a significant CSF increase (t=2.96, pb0.004) inthe total BD sample. Similarly, the BDP and BDPD subsamplesdiffered significantly from healthy controls with respect to total

Table 3Regional CSF increases in bipolar disorder.

Patient subgroup Hemisphere Talairach co

x

Sylvian fissure BD c – –

BDP d – –

BDPD e left −47Temporal sulci BD – –

BDP right 59BDPD right 59

a Local maximum.b Whole-brain corrected for multiple comparisons (FWE) at the voxel level.c Patients satisfying DSM-IV criteria for bipolar disorder (relative to controls).d Subgroup of BD patients with psychotic features (relative to controls).e Subgroup of psychotic BD patients with a lifetime history of persecutory delusi

GM and CSF volumes (BDP GM: F=11.60, pb0.001; BDP CSF:F=6.26, pb0.015; BDPD GM: F=13.72, pb0.001; BDPD GM:F=9.10, pb0.004). In contrast, between-group comparison ofBD patient subgroups yielded no significant differences in globalGM, WM or CSF. None of the performed patient-control orpatient-patient group comparisons yielded a significant volu-metric difference for TIV or total WM volume.

3.3. Regional brain morphometric results

The regional morphometric findings of this study arepresented in Tables 2 and 3. Over all, the observed differences

ordinates a Cluster size t-value a p-value a, b

y z

– – – – –

– – – – –

15 −7 760 6.0 0.003– – – – –

−13 −5 12 5.1 0.027−13 −4 32 5.5 0.017

ons (relative to controls).

58 H. Tost et al. / Journal of Affective Disorders 120 (2010) 54–61

in brain tissue concentrations were largely dependent on theBD subsample examined. Analysis of the total BD samplerelative to healthy controls yielded only one area of decreasedgray matter concentration in the left middle temporal lobe(BA 21). The BDP patient subsample showed a larger bilateralGM decrease in the middle temporal lobe (BA 21), in additionto foci in the superior temporal (BA 22), cingulate (BA 31),and lateral prefrontal cortex (BA 47). The most prominentresult in terms of the number, extent and significance ofaffected areas, however, was obtained for the BDPD sub-sample relative to healthy controls. Here, a widespreadpattern of prefrontal and temporal GM density decreaseinvolving the superior, middle and inferior temporal lobe, thedorsolateral and ventrolateral PFC, and the cingulate cortex(BAs 9, 20–22, 31, and 44–47, see Fig. 1) was seen. A similarbut inverse pattern was demonstrated for the CSF partition,where a significant CSF increase was observed for theadjacent outer ventricular sulci of the prefrontal and temporallobes. Again, this pattern was particularly pronounced in theBDPD subgroup (see Table 3 and Fig. 1). Within the BD patientgroup, comparison of BDPD patients and BD patients withoutpersecutory delusions revealed a significant GM decrease inthe left DLPFC (Talairach coordinates −40, 52, 14, t=4.84,kE=139) and left MedPFC (−12, 32, 42, t=4.45, kE=71) forthe former group (see Fig. 2). None of the performed patient-control or patient–patient group comparisons yielded asignificant difference in regional WM concentration.

4. Discussion

Bipolar disorder is a heterogeneous psychiatric conditionthat shows substantial clinical, neurobiological and genetic

Fig. 1. Statistical parametric maps of regionally reduced gray matter concentration (patient subgroups compared to healthy controls. (A) DSM-IV bipolar disorder (BD),with a history of persecutory delusions (BDPD). All illustrations are thresholded at

overlap with other diagnostic entities, in particular schizo-phrenia. Here, we investigated whether a clinical phenotypicmarker identified in prior genetic association studies of BDand SCZ also proves useful in delineating a meaningfuldisorder subgroup on a different phenotypic level, that ofstructural neuroimaging. The presented findings suggest thatwithin the DSM-IV category “Bipolar Disorder”, the patientsubgroup with persecutory delusions is characterized by acircumscribed pattern of prefrontal and temporal gray matterdeficits. Consistent with the substantial clinical overlapbetween BDPD and SCZ is the fact that similar brain structuralanomalies have only occasionally been reported in the BDliterature (Moorhead et al., 2007), but are awell-known brainmorphometric marker in SCZ (Honea et al., 2005). Further-more, compared to BD patients without persecutory delu-sions mood-incongruent psychotic symptoms, the BDPDsubgroup showed a significant reduction of GM in thedorsolateral and medial PFC, areas that have previouslybeen shown to be associated with the development ofpersecutory ideation in SCZ (Blackwood et al., 2004; Sabriet al., 1997).

4.1. Neurobiological implications

Due to its non-invasive nature, VBM yields only indirectmeasures of brain structural integrity and cannot determinethe pathophysiological background of the observed groupdifferences. However, since group differences in age, genderor TIV cannot be responsible for the observed changes inBDPD, the inverse pattern of a significant prefrontal-temporalGM decrease and significant increase of adjacent sulcalCSF probably reflects a subtle brain atrophic process. This

left panel) and increased cerebrospinal fluid (right panel) in bipolar disorder(B) Patient subgroup with psychotic symptoms (BDP), and (C) BDP subgroupa significance level of pb0.0001 (uncorrected) for presentation purposes.

Fig. 2. Significant decrease in gray matter concentration in the dorsolateralprefrontal cortex (DLPFC) of bipolar disorder patients with persecutorydelusions (bipolar PD+) relative to bipolar disorder patients withoutpersecutory delusions (bipolar PD−).

59H. Tost et al. / Journal of Affective Disorders 120 (2010) 54–61

outcome does not accord well with contemporary diseasemodels of BD that assume abnormalities of cell proliferation,neuronal hypertrophy, and pruning deficits as being keypathophysiological mechanisms (Brambilla et al., 2005;Deicken et al., 2001). In contrast, decremental structuralchanges in the temporal and prefrontal lobe of SCZ patientshave repeatedly been reported. As the structural and func-tional integrity of the prefrontal lobe is tightly linked toworking memory and executive functions, it appears plau-sible, that the pronounced structural anomalies in the DLPFCand MedPFC might relate to the prefrontal cognitive deficitsthat have been previously reported in psychotic BD patients(Bora et al., 2007; Glahn et al., 2007). Further neuroimagingstudies are necessary to address this question.

4.2. Methodological considerations

Voxel-based morphometry is a fully automated methodthat allows the unbiased analysis of structural differences ofthe whole brain on a voxel-by-voxel basis (Ashburner andFriston, 2000). Although the approach is generally consideredsuperior to conventional region of interest (ROI) analyses interms of regional sensitivity and reliability, the statistical

power of VBM in the detection of circumscribed structuralchanges depends on several factors. Firstly, due to thesubstantial anatomical variability between humans, allstructural images are spatially normalized into the samestereotactic space, a step that is subject to misalignment error,especially in small subcortical regions. Secondly, the like-lihood of detecting statistical differences in areas of a certainsize is influenced by the width of the smoothing kernelemployed (matched filter theorem) (Jones et al., 2005). Inthis study, we chose a larger smoothing kernel of 12 mm toensure that the main requirement of statistical parametrictesting of VBM data (i.e., normality of the residuals) was met,even in the scenario of subgroup comparisons (Viviani et al.,2007). This implies that our study has been sensitizedtowards the detection of morphological changes of the largercortical surface, while low statistical power for smallsubcortical brain structures must be assumed. This probablyexplains why we were not able to detect any of the medialtemporal lobe changes that have been previously reported inBD and SCZ, especially in the amygdala and hippocampus.

Another important question refers to the potential impactof psychotropic medication on the observed neurostructuralfindings. Previous longitudinal studies in humans andanimals have shown that chronic exposure to conventionalneuroleptics is associated with significantly smaller graymatter volumes in the cerebral cortex (Konopaske et al.,2007; Lieberman et al., 2005), while mood stabilizers andsome atypical antipsychotics seem to exert neuroprotectiveeffects (Manji et al., 1999; Nakamura et al., 2007). It isplausible that within the context of a symptom-orientatedtreatment regimen BD patients with mood-incongruentpsychotic features may receive typical neuroleptics moreoften and in higher doses than is the case for patients witheither mood-congruent psychotic features or purely affectivesymptomatology. Although the possibility of a confoundedmedication effect cannot be entirely ruled out, there areseveral key points that strongly argue against this interpreta-tion. Firstly, our patient study groups did not differ signifi-cantly with respect to their currently prescribed medicationtype. Secondly, the most prominent and reliable morpho-metric marker of long-term neuroleptic exposure, significantvolumetric expansion of the basal ganglia, was not observedin our BDPD patient group (Andersson et al., 2002; Potvinet al., 2007).

4.3. Nosological implications

On a more general level, our data support the notion thatthe current inconsistency in the results of BD neuroimaging is,at least in part, explained by the manner in which psychiatricdisease entities are currently conceptualized. The foundationsof the current nosological distinction between BD and SCZ canbe traced back to Emil Kraepelin's dichotomous classificationof dementia praecox and manic–depressive illness of 1919(Kraepelin, 1919). Although nearly a century has passed sincethe publication of this milestone work, the operationalcriteria implemented inmodern classification systems remainentirely descriptive, a fact explained by the lack of reliable andvalid biomarkers for the guiding of the diagnostic process.However, biomedical research conducted by means of thesecriteria proceeds under the assumption that BD and SCZ

60 H. Tost et al. / Journal of Affective Disorders 120 (2010) 54–61

represent discrete disorder entities with differing underlyingetiologies.

Recent molecular genetics findings demonstrated anoverlap in genetic susceptibility across traditional diagnosticcategories (Abou Jamra et al., 2006; Schulze et al., 2005).These results stimulated the formulation of new pathophy-siological models for the conceptualization of BD and SCZ,viewing them rather as prototypical endpoints of a spectrumof psychiatric phenotypes sharing a common set of suscept-ibility genes (Craddock et al., 2006). The neurostructuraldeficits observed in our study reflect this view, and suggestthat some of the phenotypes that are currently subsumedunder the diagnostic entity BD share relevant features withSCZ patients on the level of the intermediate neurobiologicalsubstrate. Previous electrophysiological studies e.g. suggestthat the shared impairments of the P50 auditory evokedpotential in BD and SCZ reflects the shared vulnerability topsychosis (Bora et al., 2008; Olincy and Martin, 2005). Futuremultimodal research approaches using electrophysiological,neuroimaging and cognitive measures are warranted as theymay contribute substantially to the current nosologicaldiscussion.

4.4. Limitations

Several limitations of this study warrant consideration.First, although the overall sample size of BD patients isreasonable for a neuroimaging approach, the examined BDPDpatient subsample is rather small. Among others, thisimplicates that the influence of other, even rarer psycho-pathological phenomena in our BD group could not beexamined due to limitations in statistical power (i.e.,influence of hallucinations, negative symptoms). Second,although our findings suggest that the development ofpersecutory delusions in BD indicates a more severe, schizo-phrenia-like phenotype of the illness associated with fronto-temporal brain structural anomalies, our finding do notnecessarily indicate that these findings are exclusive forBDPD patients. Future studies are necessary to explore themorphometric profile of other psychotic BD subgroups,especially psychotic BD subjects with a history of mood-incongruent delusions (but no history of mood-congruentdelusions) in comparison to BD subjects with a history ofmood-congruent delusions (but no history of mood-incon-gruent delusions).

5. Conclusions

The structural imaging results of the present study suggestthat the precise phenomenological characterization andrefined definition of patient samples is a crucial step inovercoming the profound inconsistency of results observed inBD neuroimaging research. Our findings support the conceptthat the BDPD patients belong to a biologically meaningfulsubgroup within the affective-psychotic spectrum of mentaldisorders, a subgroup which shares relevant genetic, neuro-biological and clinical features with SCZ patients. In accor-dance with previous genetic studies in the field, our datacontradict the dichotomy of “functional psychoses”, andquestion the biological validity of current psychiatric nosol-ogy. Further neuroimaging andmolecular genetics studies are

necessary in order to clarify the shared pathophysiologicalfactors involved in the etiology of BD and SCZ disorderphenotypes and examine whether the described morpho-metric findings are specific for BDPD patients.

Role of funding sourceThis work was supported by grants from the National German Genome

Research Network (NGFN) of the Federal Ministry of Education and Research(BMBF). NGFN and BMBF had no further role in study design; in thecollection, analysis and interpretation of data; in the writing of the report;and in the decision to submit the paper for publication.

Conflict of interestAll authors declare that they have no conflicts of interest.

Acknowledgment

The authors thank Elfriede Luebbeke for her invaluableassistance during MRI data acquisition, Traute Demirakça forher valuable methodological input, Monika Deschner for hersupport in patient recruitment and diagnostic rating, andMarkus Nöthen and Shabnam Hakimi for their helpfulcomments and critical reading of the manuscript.

References

Abou Jamra, R., Schmael, C., Cichon, S., Rietschel, M., Schumacher, J., Nothen,M.M., 2006. The G72/G30 gene locus in psychiatric disorders: a challengeto diagnostic boundaries? Schizophr. Bull. 32, 599–608.

Adler, C.M., DelBello, M.P., Jarvis, K., Levine, A., Adams, J., Strakowski, S.M.,2007. Voxel-based study of structural changes in first-episode patientswith bipolar disorder. Biol. Psychiatry 61, 776–781.

Andersson, C., Hamer, R.M., Lawler, C.P., Mailman, R.B., Lieberman, J.A., 2002.Striatal volume changes in the rat following long-term administration oftypical and atypical antipsychotic drugs. Neuropsychopharmacology 27,143–151.

Ashburner, J., Friston, K.J., 2000. Voxel-based morphometry — the methods.NeuroImage 11, 805–821.

Black, D.W., Nasrallah, A., 1989. Hallucinations and delusions in 1,715 patientswith unipolar and bipolar affective disorders. Psychopathology 22,28–34.

Blackwood, N.J., Bentall, R.P., Ffytche, D.H., Simmons, A., Murray, R.M.,Howard, R.J., 2004. Persecutory delusions and the determination of self-relevance: an fMRI investigation. Psychol. Med. 34, 591–596.

Bora, E., Vahip, S., Akdeniz, F., Gonul, A.S., Eryavuz, A., Ogut, M., Alkan, M.,2007. The effect of previous psychotic mood episodes on cognitiveimpairment in euthymic bipolar patients. Bipolar Disord. 9, 468–477.

Bora, E., Yucel, M., Fornito, A., Berk, M., Pantelis, C., 2008. Major psychoseswith mixed psychotic and mood symptoms: are mixed psychosesassociated with different neurobiological markers? Acta Psychiatr.Scand. 118, 172–187.

Brambilla, P., Glahn, D.C., Balestrieri, M., Soares, J.C., 2005. Magneticresonance findings in bipolar disorder. Psychiatr. Clin. North Am. 28,443–467.

Corlett, P.R., Murray, G.K., Honey, G.D., Aitken, M.R., Shanks, D.R., Robbins, T.W.,Bullmore, E.T., Dickinson, A., Fletcher, P.C., 2007. Disrupted prediction-errorsignal in psychosis: evidence for an associative account of delusions. Brain130, 2387–2400.

Craddock, N., O'Donovan, M.C., Owen,M.J., 2006. Genes for schizophrenia andbipolar disorder? Implications for psychiatric nosology. Schizophr. Bull.32, 9–16.

Deicken, R.F., Eliaz, Y., Feiwell, R., Schuff, N., 2001. Increased thalamic N-acetylaspartate in male patients with familial bipolar I disorder. PsychiatryRes. 106, 35–45.

First, M.B., Spitzer, R.L., Gibbon, M., Williams, J.B.W., 2002. Structured ClinicalInterview for DSM-IV-TR Axis I Disorders, Patient Edition (SCID-I/P).Biometrics Research Department, New York State Psychiatric Institute,New York.

Friston, K.J., Holmes, A.P., Worsley, K.J., Poline, J.P., Frith, C.D., Frackowiak, R.S.,1995. Statistical parametric maps in functional imaging: a general linearmodel approach. Hum. Brain Mapp. 2, 189–210.

61H. Tost et al. / Journal of Affective Disorders 120 (2010) 54–61

Glahn, D.C., Bearden, C.E., Barguil, M., Barrett, J., Reichenberg, A., Bowden, C.L.,Soares, J.C., Velligan, D.I., 2007. The neurocognitive signature of psychoticbipolar disorder. Biol. Psychiatry 62, 910–916.

Goes, F.S., Zandi, P.P.,Miao,K.,McMahon, F.J., Steele, J.,Willour, V.L.,Mackinnon,D.F.,Mondimore, F.M., Schweizer, B., Nurnberger Jr., J.I., Rice, J.P., Scheftner, W.,Coryell,W.,Berrettini,W.H.,Kelsoe, J.R., Byerley,W.,Murphy,D.L.,Gershon,E.S.,BipolarDisorder Phenome,G., Depaulo Jr., J.R.,McInnis,M.G., Potash, J.B., 2007.Mood-incongruentpsychotic features in bipolardisorder: familial aggregationand suggestive linkage to 2p11-q14 and 13q21-33. Am. J. Psychiatry 164,236–247.

Good, C.D., Johnsrude, I.S., Ashburner, J., Henson, R.N., Friston, K.J.,Frackowiak, R.S., 2001. A voxel-based morphometric study of ageing in465 normal adult human brains. NeuroImage 14, 21–36.

Green, E.K., Raybould, R., Macgregor, S., Gordon-Smith, K., Heron, J., Hyde, S.,Grozeva, D., Hamshere, M., Williams, N., Owen, M.J., O'Donovan, M.C.,Jones, L., Jones, I., Kirov, G., Craddock, N., 2005. Operation of theschizophrenia susceptibility gene, neuregulin 1, across traditionaldiagnostic boundaries to increase risk for bipolar disorder. Arch. Gen.Psychiatry 62, 642–648.

Harrison, P.J., 1999. The neuropathology of schizophrenia. A critical review ofthe data and their interpretation. Brain 122 (Pt 4), 593–624.

Harrow, M., Grossman, L.S., Herbener, E.S., Davies, E.W., 2000. Ten-yearoutcome: patients with schizoaffective disorders, schizophrenia, affec-tive disorders and mood-incongruent psychotic symptoms. Br. J.Psychiatry 177, 421–426.

Honea, R., Crow, T.J., Passingham, D., Mackay, C.E., 2005. Regional deficits inbrain volume in schizophrenia: a meta-analysis of voxel-based morpho-metry studies. Am. J. Psychiatry 162, 2233–2245.

Jones, D.K., Symms, M.R., Cercignani, M., Howard, R.J., 2005. The effect of filtersize on VBM analyses of DT-MRI data. NeuroImage 26, 546–554.

Konopaske, G.T., Dorph-Petersen, K.A., Pierri, J.N., Wu, Q., Sampson, A.R.,Lewis, D.A., 2007. Effect of chronic exposure to antipsychotic medicationon cell numbers in the parietal cortex of macaque monkeys. Neuropsy-chopharmacology 32, 1216–1223.

Kraepelin, E., 1919. Dementia praecox and paraphrenia. In: Robertson, G.M.,Livingstone, S. (Eds.), Translated by R. Mary Barclay from the eighthGerman edition of the “Text-Book of Psychiatry”, vol. III, part II. ChicagoMedical Book Company, Chicago.

Lehrl, S., 2005. Mehrfachwahl-Wortschatz-Intelligenztest MWT-B. SpittaVerlag GmbH.

Lieberman, J.A., Tollefson, G.D., Charles, C., Zipursky, R., Sharma, T., Kahn, R.S.,Keefe, R.S., Green, A.I., Gur, R.E., McEvoy, J., Perkins, D., Hamer, R.M., Gu, H.,Tohen, M., 2005. Antipsychotic drug effects on brain morphology in first-episode psychosis. Arch. Gen. Psychiatry 62, 361–370.

Maldjian, J.A., Laurienti, P.J., Kraft, R.A., Burdette, J.H., 2003. An automatedmethod for neuroanatomic and cytoarchitectonic atlas-based interroga-tion of fMRI data sets. NeuroImage 19, 1233–1239.

Manji, H.K., Moore, G.J., Chen, G., 1999. Lithium at 50: have the neuropro-tective effects of this unique cation been overlooked? Biol. Psychiatry 46,929–940.

McDonald, C., Zanelli, J., Rabe-Hesketh, S., Ellison-Wright, I., Sham, P.,Kalidindi, S., Murray, R.M., Kennedy, N., 2004. Meta-analysis of magneticresonance imaging brain morphometry studies in bipolar disorder. Biol.Psychiatry 56, 411–417.

McDonald, C., Marshall, N., Sham, P.C., Bullmore, E.T., Schulze, K., Chapple, B.,Bramon, E., Filbey, F., Quraishi, S., Walshe, M., Murray, R.M., 2006.

Regional brain morphometry in patients with schizophrenia or bipolardisorder and their unaffected relatives. Am. J. Psychiatry 163, 478–487.

McGuffin, P., Farmer, A., Harvey, I., 1991. A polydiagnostic application ofoperational criteria in studies of psychotic illness. Development andreliability of the OPCRIT system. Arch. Gen. Psychiatry 48, 764–770.

Moorhead, T.W., McKirdy, J., Sussmann, J.E., Hall, J., Lawrie, S.M., Johnstone, E.C.,McIntosh, A.M., 2007. Progressive gray matter loss in patients with bipolardisorder. Biol. Psychiatry 62, 894–900.

Nakamura, M., Salisbury, D.F., Hirayasu, Y., Bouix, S., Pohl, K.M., Yoshida, T.,Koo, M.S., Shenton, M.E., McCarley, R.W., 2007. Neocortical gray mattervolume in first-episode schizophrenia and first-episode affectivepsychosis: a cross-sectional and longitudinal MRI study. Biol. Psychiatry62, 773–783.

Olincy, A., Martin, L., 2005. Diminished suppression of the P50 auditoryevoked potential in bipolar disorder subjects with a history of psychosis.Am. J. Psychiatry 162, 43–49.

Ongur, D., Price, J.L., 2000. The organization of networks within the orbitaland medial prefrontal cortex of rats, monkeys and humans. Cereb. Cortex10, 206–219.

Potvin, S., Mancini-Marie, A., Fahim, C., Mensour, B., Levesque, J., Karama, S.,Beauregard, M., Rompre, P.P., Stip, E., 2007. Increased striatal gray matterdensities in patients with schizophrenia and substance use disorder: avoxel-based morphometry study. Psychiatry Res. 154, 275–279.

Regeer, E.J., ten Have, M., Rosso, M.L., Hakkaart-van Roijen, L., Vollebergh, W.,Nolen, W.A., 2004. Prevalence of bipolar disorder in the generalpopulation: a reappraisal study of the Netherlands Mental Health Surveyand Incidence Study. Acta Psychiatr. Scand. 110, 374–382.

Sabri, O., Erkwoh, R., Schreckenberger, M., Cremerius, U., Owega, A.,Dickmann, C., Schulz, G., Zimny, M., Sass, H., Buell, U., 1997. Alteredrelationships between rCBF in different brain regions of never-treatedschizophrenics. Nuklearmediziner 36, 194–201.

Schulze, T.G., Ohlraun, S., Czerski, P.M., Schumacher, J., Kassem, L., Deschner, M.,Gross, M., Tullius, M., Heidmann, V., Kovalenko, S., Jamra, R.A., Becker, T.,Leszczynska-Rodziewicz, A., Hauser, J., Illig, T., Klopp, N.,Wellek, S., Cichon, S.,Henn, F.A.,McMahon, F.J., Maier,W., Propping, P., Nothen,M.M., Rietschel,M.,2005. Genotype-phenotype studies in bipolar disorder showing associationbetween theDAOA/G30 locus andpersecutorydelusions: afirst step towardamolecular genetic classification of psychiatric phenotypes. Am. J. Psychiatry162, 2101–2108.

Strakowski, S.M., Delbello, M.P., Adler, C.M., 2005. The functional neuroa-natomy of bipolar disorder: a review of neuroimaging findings. Mol.Psychiatry 10, 105–116.

van Haren, N.E., Hulshoff Pol, H.E., Schnack, H.G., Cahn, W., Mandl, R.C.,Collins, D.L., Evans, A.C., Kahn, R.S., 2007. Focal gray matter changes inschizophrenia across the course of the illness: a 5-year follow-up study.Neuropsychopharmacology 32, 2057–2066.

Viviani, R., Beschoner, P., Ehrhard, K., Schmitz, B., Thone, J., 2007. Non-normalityand transformations of random fields, with an application to voxel-basedmorphometry. NeuroImage 35, 121–130.

Wright, I.C., Rabe-Hesketh, S., Woodruff, P.W., David, A.S., Murray, R.M.,Bullmore, E.T., 2000. Meta-analysis of regional brain volumes in schizo-phrenia. Am. J. Psychiatry 157, 16–25.

Young, R.C., Biggs, J.T., Ziegler, V.E., Meyer, D.A., 2000. Young mania ratingscale. Handbook of Psychiatric Measures. American Psychiatric Associa-tion, Washington, DC, pp. 540–542.