Research paper
The impact of psychosis on brain anatomy in bipolar disorder: A structural MRI study

https://doi.org/10.1016/j.jad.2017.11.092Get rights and content

Highlights

  • The impact of psychosis on brain anatomy in bipolar disorder is underestimated.

  • We study the neuroanatomical correlates of psychosis in bipolar disorder patients.

  • Gray matter volume alterations are enhanced in psychotic patients.

  • Results confirm the importance of classifying bipolar patients based on psychosis.

Abstract

Background

Bipolar disorder (BD) is a major psychiatric illness characterized by heterogeneous symptoms including psychotic features. Up until now, neuroimaging studies investigating cerebral morphology in patients with BD have underestimated the potential impact of psychosis on brain anatomy in BD patients. In this regard, psychotic and non-psychotic BD may represent biologically different subtypes of the disorder, being possibly associated with specific cerebral features.

Methods

In the present study, magnetic resonance imaging (MRI) at 3T was used to identify the neuroanatomical correlates of psychosis in an International sample of BD patients. A large sample of structural MRI data from healthy subjects (HC) and BD patients was collected across two research centers. Voxel based morphometry was used to compare gray matter (GM) volume among psychotic and non-psychotic BD patients and HC.

Results

We found specific structural alterations in the two patient groups, more extended in the psychotic sample. Psychotic patients showed GM volume deficits in left frontal cortex compared to HC, and in right temporo-parietal cortex compared to both HC and non-psychotic patients (p < 0.001, > 100 voxels). Psychotic patients also exhibited enhanced age-related GM volume deficits in a set of subcortical and cortical regions.

Limitations

The integration of multiple datasets may have affected the results.

Conclusions

Overall, our results confirm the importance of classifying BD based on psychosis. The knowledge of the neuronal bases of psychotic symptomatology in BD can provide a more comprehensive picture of the determinants of BD, in the light of the continuum characteristic of major psychoses.

Introduction

Bipolar disorder (BD) is a severe and chronic neuropsychiatric condition that affects mood regulation, being characterized by recurring depressive, manic or hypomanic, mixed and euthymic phases. This complex illness is associated with heterogeneous clinical features (Phillips and Kupfer, 2013) and variable patterns of cognitive deficits (Martino et al., 2008), ultimately influencing patients’ clinical and functional outcome.

In the search for objective neurobiological markers of BD, advanced neuroimaging techniques have been increasingly employed. Indeed, the neural correlates of BD have been extensively investigated using Magnetic Resonance Imaging (MRI), with results showing structural alterations in BD patients compared to healthy controls (HC) (Arnone et al., 2008, Ellison-Wright and Bullmore, 2010, Hallahan et al., 2011, Selvaraj et al., 2012). However, taken together, the region-based and voxel-based neuroanatomical studies on BD led to rather inconsistent results.

This inconsistency might be related to the heterogeneous clinical manifestations within the BD spectrum, which were not adequately considered by most of the studies. In this regard, a feature that is worth of consideration is the presence of psychotic symptoms, which seems to have clinical, cognitive and psychosocial implications (Coryell et al., 2001, Glahn et al., 2007). For instance, a recent work from our group has shown that lifetime psychosis in BD is related to a worse socio-demographic and clinical profile (Dell'Osso et al., 2017). Literature evidence shows that more than half of BD patients experience psychotic symptoms at least once in the lifetime, more frequently during manic episodes (Goodwin and Jamison, 2007). Among patients with BD type I, prevalence rates of psychosis of up to 68% have been reported (Keck et al., 2003); as expected, psychotic symptoms occur in a lower percentage of BD type II patients, which was estimated in a range up to 45% (Mazzarini et al., 2010).

Besides influencing the choice of the pharmacological therapy, the presence of psychotic features was found to be associated with an earlier age of onset (Schurhoff et al., 2000), greater cross-sectional symptom severity and long-term morbidity (Coryell et al., 2001), as well as increased D2 dopamine receptor density (Pearlson et al., 1995). Family studies provided consistent evidence that psychotic symptoms aggregate in BD pedigrees and further suggested that liability to psychosis can be transmitted across different diagnoses within the psychotic spectrum, supporting the hypothesis that psychotic BD is associated with “overlapping” genes between BD and schizophrenia (SCZ) (Goes et al., 2008). Indeed, over recent years, researchers investigating the genetic bases of BD overcame the traditional diagnostic category by posing attention to different phenotypes within BD, with special interest towards psychotic features (Craddock et al., 2005, Park et al., 2004, Potash et al., 2003). A gradient of polygenic risk score association across SCZ and BD indexed by the presence and nature of psychosis has been recently demonstrated, with BD patients with mood-incongruent psychotic features carrying the highest burden of SCZ risk alleles (Allardyce et al., 2017).

Psychotic and non-psychotic BD may indeed represent different subtypes of the illness, relying on specific neuroanatomical bases. Therefore, in neuroimaging studies, the integration of BD patients with and without psychosis in a single diagnostic group might cause misinterpretations and obscure subgroup characterization (Strasser et al., 2005). Taking into consideration psychotic features can ameliorate the identification of the structural markers of BD, in line with the dimensional framework of Research Domain Criteria (RDoC) (Insel et al., 2010).

The largest study on cortical thickness and cortical surface area in BD, recently published from the ENIGMA group, showed an association between lifetime psychosis and deficits of cortical surface area in selective frontal and temporal regions (Hibar et al., 2017). Specifically, adolescents with psychotic BD showed deficits in right caudal anterior cingulate cortex and left inferior temporal gyrus compared to adolescents with non-psychotic BD, whereas adults with psychotic BD were characterized by deficits in right frontal pole compared to non-psychotic BD patients in the same age range.

Mixed results came from the investigation of gray matter (GM) volume in these patients. Indeed, there is just a handful of MRI studies that compared psychotic and non-psychotic BD patients in terms of GM volume (Chen et al., 2007, Ekman et al., 2017, Keramatian et al., 2016, Laidi et al., 2015, Mamah et al., 2016, Neves et al., 2016, Strasser et al., 2005, Womer et al., 2014). As recently summarized by our group (Maggioni et al., 2017), the findings of these studies suggest that psychotic and non-psychotic BD patients show specific volume alterations, enhanced in the psychotic group, when compared to controls. The results from the direct comparisons between the two BD subtypes suggest that psychosis is associated with smaller cortical volume, especially in the frontal area, and larger volume in selective regions of the basal ganglia. Additionally, within the psychotic BD spectrum, mood-incongruent features have been recently associated with enhanced GM volume alterations, in regions that seem to be involved in SCZ (Keramatian et al., 2016).

The above-mentioned results support the hypothesis that BD with psychotic features may represent a neurobiologically homogeneous subphenotype of the disorder; however, firm conclusions cannot be drawn given the limited number of studies, smaller sample sizes and inconsistencies in the results. The present work was undertaken to gain more robust understanding of the neuroanatomical underpinnings of psychosis. By adopting an unbiased exploratory perspective, we investigated whole-brain GM volume in a large structural MRI dataset of HC and patients with psychotic and non-psychotic BD, which was collected across different Research Centers, the IRCCS Fondazione Ca’ Granda Ospedale Maggiore Policlinico in Milan, Italy, and the University of British Columbia affiliated Hospitals in Vancouver, Canada. The Canadian dataset was obtained from the Systematic Treatment Optimization Program for Early Mania (STOP-EM), whose protocol is described elsewhere (Yatham et al., 2009). Structural and functional MRI studies on subsets of the multicenter dataset have been recently published (Dell'Osso et al., 2015, Keramatian et al., 2016).

Section snippets

Subjects

One-hundred seventy-one subjects (114 BD patients and 57 HC) were recruited across the two research centers. Preliminarily to the MRI acquisition, all the subjects signed a written informed consent to the protocol, in accordance with the Declaration of Helsinki and the local ethical committee guidelines. After a quality check procedure, 164 subjects were selected for the analysis. The analysis of the entire dataset (Dataset1, n = 164) was accompanied by the analysis of a reduced dataset of

Demographic and clinical information

The demographic and clinical details of Dataset1 are listed in Table 1. The dataset was composed of 56 HC (29 males, range 18–62 years, 25.3 ± 7.37 years), 46 non-psychotic BD patients (23 males, range 16–65 years, 33.67 ± 11.75 years) and 62 psychotic BD patients (27 males, range 17–51 years, 27.27 ± 8.70 years). Sex ratio (M/F) was 0.51 in HC, 0.5 in non-psychotic BD and 0.44 in psychotic BD (χ2 = 0.89, not significant). The ANOVA showed significant age differences across groups (F = 11.04, p

Discussion

The present work focused on the GM volume analysis of a large structural MRI dataset composed of BD patients, with and without lifetime psychosis, and healthy reference subjects. Our main interest was to assess whether psychotic symptoms in BD are associated with a specific neurobiological architecture, ultimately verifying the importance to classify BD based on the psychotic dimension.

To provide robust evidence on the GM volume correlates of psychotic and non-psychotic BD, overcoming the low

Conclusions

Despite the pathophysiology of BD has been largely investigated in the last decades, the neuroanatomical underpinnings of this complex disorder are still debated. The variety of clinical manifestations in BD may have contributed to the heterogeneity of neuroimaging findings on BD patients (Ellison-Wright and Bullmore, 2010). A relevant clinical variable is represented by psychotic symptoms, which have an impact on the patients’ clinical outcome and may have specific cerebral features.

In the

Acknowledgement

Grant support of Dr. Maggioni was provided by the European Union's Seventh Framework Programme for research, technological development and demonstration under Grant agreement no. 602450 (IMAGEMEND Project) and by the Fondazione Cariplo (grant n° 2016-0908). Prof. Paolo Brambilla was partly supported by the Italian Ministry of Health (RF-2011‐02352308).

Role of the funding source

Grant support of Dr. Maggioni was provided by the European Union's Seventh Framework Programme for research, technological development and demonstration under Grant agreement no. 602450 (IMAGEMEND Project) and by the Fondazione Cariplo (grant n° 2016-0908). Prof. Paolo Brambilla was partly supported by the Italian Ministry of Health (RF-2011‐02352308).

Ethical statement

This research involves human participants. Preliminarily to the MRI study, all the subjects signed a written informed consent to the protocol, in

References (59)

  • D. Mamah et al.

    Subcortical neuromorphometry in schizophrenia spectrum and bipolar disorders

    Neuroimage Clin.

    (2016)
  • D.J. Martino et al.

    Heterogeneity in cognitive functioning among patients with bipolar disorder

    J. Affect. Disord.

    (2008)
  • L. Mazzarini et al.

    Psychotic versus non-psychotic bipolar II disorder

    J. Affect. Disord.

    (2010)
  • S.A. Mitelman et al.

    Volume of the cingulate and outcome in schizophrenia

    Schizophr. Res.

    (2005)
  • M.L. Phillips et al.

    Bipolar disorder diagnosis: challenges and future directions

    Lancet

    (2013)
  • F. Schurhoff et al.

    Early and late onset bipolar disorders: two different forms of manic-depressive illness?

    J. Affect. Disord.

    (2000)
  • L. Squarcina et al.

    Classification of first-episode psychosis in a large cohort of patients using support vector machine and multiple kernel learning techniques

    Neuroimage

    (2017)
  • L. Squarcina et al.

    The use of dynamic susceptibility contrast (DSC) MRI to automatically classify patients with first episode psychosis

    Schizophr. Res.

    (2015)
  • H.C. Strasser et al.

    Hippocampal and ventricular volumes in psychotic and nonpsychotic bipolar patients compared with schizophrenia patients and community control subjects: a pilot study

    Biol. Psychiatry

    (2005)
  • F.Y. Womer et al.

    Basal ganglia and thalamic morphology in schizophrenia and bipolar disorder

    Psychiatry Res.

    (2014)
  • J. Allardyce et al.

    Psychosis and the level of mood incongruence in bipolar disorder are related to genetic liability for schizophrenia

    bioRxiv

    (2017)
  • D. Arnone et al.

    Meta-analysis of magnetic resonance imaging studies of the corpus callosum in bipolar disorder

    Acta Psychiatr. Scand.

    (2008)
  • American Psychiatric Association

    DSM-IV-TR: Diagnostic and Statistical Manual of Mental Disorders, Text Revision

    (2000)
  • C.E. Bearden et al.

    The neuropsychology and neuroanatomy of bipolar affective disorder: a critical review

    Bipolar Disord.

    (2001)
  • M. Bellani et al.

    Laterality effects in schizophrenia and bipolar disorder

    Exp. Brain Res.

    (2010)
  • P. Brambilla et al.

    White matter connectivity in bipolar disorder

    Int. Rev. Psychiatry

    (2009)
  • P. Brambilla et al.

    The role of white matter for the pathophysiology of schizophrenia

    Int. Rev. Psychiatry

    (2007)
  • M.P. Caligiuri et al.

    A functional magnetic resonance imaging study of cortical asymmetry in bipolar disorder

    Bipolar Disord.

    (2004)
  • X. Chen et al.

    Regional gray matter changes in bipolar disorder: a voxel-based morphometric study

    Aust. N. Z. J. Psychiatry

    (2007)
  • Cited by (29)

    • Large-scale network abnormality in bipolar disorder: A multimodal meta-analysis of resting-state functional and structural magnetic resonance imaging studies

      2021, Journal of Affective Disorders
      Citation Excerpt :

      In recent years, brain imaging and neuropathological studies have demonstrated structural and functional brain changes in BD. Such changes have been greatly advanced by attempts to define the distinct brain regions, their interregional connectivity, and their specific function and/or physiology which affect the large-scale brain systems (Altamura et al., 2018; Brady et al., 2016; Chen et al., 2019; Gong et al., 2019; Luo et al., 2018; Sani et al., 2016). Among these studies, resting-state functional magnetic resonance imaging (rs-fMRI), in particular, the seed-based resting-state functional connectivity (rs-FC) has primary advantages over other methods.

    • Brain structural alterations in pediatric bipolar disorder patients with and without psychotic symptoms

      2021, Journal of Affective Disorders
      Citation Excerpt :

      Our results in PBD patients with psychotic features were in line with prior evidence which also confirmed extensively distributed GM volume/density deficits in adults with psychotic BD compared with HC (Wang et al., 2018). Using the whole-brain approach, although not consistent, an overwhelming number of studies showed altered GMVs in the bilateral PFC involving IFG (Altamura et al., 2017), left PG (Altamura et al., 2018), bilateral STG (Cui et al., 2011), as well as subcortical regions, including the bilateral parahippocampus (Ivleva et al., 2013), and bilateral putamen (Chen et al., 2012) in adults with psychotic BD relative to HCs. It was interesting that genetic studies had indicated that the two major, classically distinct forms of mental disorders, schizophrenia and BD, might share causative factors in common (Moskvina et al., 2009).

    • The clinical relevance of gray matter atrophy and microstructural brain changes across the psychosis continuum

      2021, Schizophrenia Research
      Citation Excerpt :

      Thus, one of these abnormalities, or a combination of them, could potentially classify PSD along a continuum of symptom burden or cognitive impairment. It has been well established that PSD exhibit atrophy, including larger ventricles, reduced gray matter volume (potentially driven by cortical thinning; Rimol et al., 2012), and abnormal white matter connections, with the greatest atrophy occurring in fronto-temporal lobes (Bakhshi and Chance, 2015; Birur et al., 2017; Maggioni et al., 2017; Altamura et al., 2018; Vita et al., 2012; Ivleva et al., 2013; Gupta et al., 2015; van Erp et al., 2018). Decreased cortical volume in PSD is thought to be due to reduced neurites (axons and dendrites), neuronal number/size, neuropil (space between cells; i.e., reduced neuropil hypothesis; Selemon and Goldman-Rakic, 1999; Nazeri et al., 2020), and/or abnormal myelination, potentially supporting an underlying progressive neurodevelopmental abnormality rather than neurodegeneration per se due to the absence of gliosis or gross inflammatory processes (Bakhshi and Chance, 2015; Harrison et al., 2018; Iritani, 2013).

    • Specific white matter connectomic changes in schizophrenia compared with psychotic bipolar disorder

      2021, Asian Journal of Psychiatry
      Citation Excerpt :

      The number of studies of first episode patients with limited medication effect are inadequate. In addition, the alterations of brain network properties may differ between BD patients with and without psychotic symptoms (Altamura et al., 2018). Thus, studying BD patients with psychotic symptoms may be more applicable.

    View all citing articles on Scopus
    1

    The two authors contributed equally to the study.

    View full text